6 This file is a HOWTO for Wireshark developers. It describes how to start coding
7 a Wireshark protocol dissector and the use of some of the important functions
10 This file is compiled to give in depth information on Wireshark.
11 It is by no means all inclusive and complete. Please feel free to send
12 remarks and patches to the developer mailing list.
16 Before starting to develop a new dissector, a "running" Wireshark build
17 environment is required - there's no such thing as a standalone "dissector
20 How to setup such an environment is platform dependent; detailed information
21 about these steps can be found in the "Developer's Guide" (available from:
22 http://www.wireshark.org) and in the INSTALL and README files of the sources
25 0.1. General README files.
27 You'll find additional information in the following README files:
29 - README.capture - the capture engine internals
30 - README.design - Wireshark software design - incomplete
31 - README.developer - this file
32 - README.display_filter - Display Filter Engine
33 - README.idl2wrs - CORBA IDL converter
34 - README.packaging - how to distribute a software package containing WS
35 - README.regression - regression testing of WS and TS
36 - README.stats_tree - a tree statistics counting specific packets
37 - README.tapping - "tap" a dissector to get protocol specific events
38 - README.xml-output - how to work with the PDML exported output
39 - wiretap/README.developer - how to add additional capture file types to
42 0.2. Dissector related README files.
44 You'll find additional dissector related information in the following README
47 - README.binarytrees - fast access to large data collections
48 - README.heuristic - what are heuristic dissectors and how to write them
49 - README.malloc - how to obtain "memory leak free" memory
50 - README.plugins - how to "pluginize" a dissector
51 - README.python - writing a dissector in PYTHON.
52 - README.request_response_tracking - how to track req./resp. times and such
56 James Coe <jammer[AT]cin.net>
57 Gilbert Ramirez <gram[AT]alumni.rice.edu>
58 Jeff Foster <jfoste[AT]woodward.com>
59 Olivier Abad <oabad[AT]cybercable.fr>
60 Laurent Deniel <laurent.deniel[AT]free.fr>
61 Gerald Combs <gerald[AT]wireshark.org>
62 Guy Harris <guy[AT]alum.mit.edu>
63 Ulf Lamping <ulf.lamping[AT]web.de>
65 1. Setting up your protocol dissector code.
67 This section provides skeleton code for a protocol dissector. It also explains
68 the basic functions needed to enter values in the traffic summary columns,
69 add to the protocol tree, and work with registered header fields.
75 Wireshark runs on many platforms, and can be compiled with a number of
76 different compilers; here are some rules for writing code that will work
77 on multiple platforms.
79 Don't use C++-style comments (comments beginning with "//" and running
80 to the end of the line); Wireshark's dissectors are written in C, and
81 thus run through C rather than C++ compilers, and not all C compilers
82 support C++-style comments (GCC does, but IBM's C compiler for AIX, for
83 example, doesn't do so by default).
85 In general, don't use C99 features since some C compilers used to compile
86 Wireshark don't support C99 (E.G. Microsoft C).
88 Don't initialize variables in their declaration with non-constant
89 values. Not all compilers support this. E.g. don't use
90 guint32 i = somearray[2];
96 Don't use zero-length arrays; not all compilers support them. If an
97 array would have no members, just leave it out.
99 Don't declare variables in the middle of executable code; not all C
100 compilers support that. Variables should be declared outside a
101 function, or at the beginning of a function or compound statement.
103 Don't use anonymous unions; not all compilers support them.
111 } u; /* have a name here */
114 Don't use "uchar", "u_char", "ushort", "u_short", "uint", "u_int",
115 "ulong", "u_long" or "boolean"; they aren't defined on all platforms.
116 If you want an 8-bit unsigned quantity, use "guint8"; if you want an
117 8-bit character value with the 8th bit not interpreted as a sign bit,
118 use "guchar"; if you want a 16-bit unsigned quantity, use "guint16";
119 if you want a 32-bit unsigned quantity, use "guint32"; and if you want
120 an "int-sized" unsigned quantity, use "guint"; if you want a boolean,
121 use "gboolean". Use "%d", "%u", "%x", and "%o" to print those types;
122 don't use "%ld", "%lu", "%lx", or "%lo", as longs are 64 bits long on
123 many platforms, but "guint32" is 32 bits long.
125 Don't use "long" to mean "signed 32-bit integer", and don't use
126 "unsigned long" to mean "unsigned 32-bit integer"; "long"s are 64 bits
127 long on many platforms. Use "gint32" for signed 32-bit integers and use
128 "guint32" for unsigned 32-bit integers.
130 Don't use "long" to mean "signed 64-bit integer" and don't use "unsigned
131 long" to mean "unsigned 64-bit integer"; "long"s are 32 bits long on
132 many other platforms. Don't use "long long" or "unsigned long long",
133 either, as not all platforms support them; use "gint64" or "guint64",
134 which will be defined as the appropriate types for 64-bit signed and
137 On LLP64 data model systems (notably 64-bit Windows), "int" and "long"
138 are 32 bits while "size_t" and "ptrdiff_t" are 64 bits. This means that
139 the following will generate a compiler warning:
142 i = strlen("hello, sailor"); /* Compiler warning */
144 Normally, you'd just make "i" a size_t. However, many GLib and Wireshark
145 functions won't accept a size_t on LLP64:
148 char greeting[] = "hello, sailor";
149 guint byte_after_greet;
151 i = strlen(greeting);
152 byte_after_greet = tvb_get_guint8(tvb, i); /* Compiler warning */
154 Try to use the appropriate data type when you can. When you can't, you
155 will have to cast to a compatible data type, e.g.
158 char greeting[] = "hello, sailor";
159 guint byte_after_greet;
161 i = strlen(greeting);
162 byte_after_greet = tvb_get_guint8(tvb, (gint) i); /* OK */
167 char greeting[] = "hello, sailor";
168 guint byte_after_greet;
170 i = (gint) strlen(greeting);
171 byte_after_greet = tvb_get_guint8(tvb, i); /* OK */
173 See http://www.unix.org/version2/whatsnew/lp64_wp.html for more
174 information on the sizes of common types in different data models.
176 When printing or displaying the values of 64-bit integral data types,
177 don't use "%lld", "%llu", "%llx", or "%llo" - not all platforms
178 support "%ll" for printing 64-bit integral data types. Instead, for
179 GLib routines, and routines that use them, such as all the routines in
180 Wireshark that take format arguments, use G_GINT64_MODIFIER, for example:
182 proto_tree_add_text(tree, tvb, offset, 8,
183 "Sequence Number: %" G_GINT64_MODIFIER "u",
186 When specifying an integral constant that doesn't fit in 32 bits, don't
187 use "LL" at the end of the constant - not all compilers use "LL" for
188 that. Instead, put the constant in a call to the "G_GINT64_CONSTANT()"
191 G_GINT64_CONSTANT(11644473600U)
197 Don't assume that you can scan through a va_list initialized by va_start
198 more than once without closing it with va_end and re-initalizing it with
199 va_start. This applies even if you're not scanning through it yourself,
200 but are calling a routine that scans through it, such as vfprintf() or
201 one of the routines in Wireshark that takes a format and a va_list as an
202 argument. You must do
204 va_start(ap, format);
205 call_routine1(xxx, format, ap);
207 va_start(ap, format);
208 call_routine2(xxx, format, ap);
212 va_start(ap, format);
213 call_routine1(xxx, format, ap);
214 call_routine2(xxx, format, ap);
217 Don't use a label without a statement following it. For example,
227 will not work with all compilers - you have to do
237 with some statement, even if it's a null statement, after the label.
239 Don't use "bzero()", "bcopy()", or "bcmp()"; instead, use the ANSI C
242 "memset()" (with zero as the second argument, so that it sets
243 all the bytes to zero);
245 "memcpy()" or "memmove()" (note that the first and second
246 arguments to "memcpy()" are in the reverse order to the
247 arguments to "bcopy()"; note also that "bcopy()" is typically
248 guaranteed to work on overlapping memory regions, while
249 "memcpy()" isn't, so if you may be copying from one region to a
250 region that overlaps it, use "memmove()", not "memcpy()" - but
251 "memcpy()" might be faster as a result of not guaranteeing
252 correct operation on overlapping memory regions);
254 and "memcmp()" (note that "memcmp()" returns 0, 1, or -1, doing
255 an ordered comparison, rather than just returning 0 for "equal"
256 and 1 for "not equal", as "bcmp()" does).
258 Not all platforms necessarily have "bzero()"/"bcopy()"/"bcmp()", and
259 those that do might not declare them in the header file on which they're
260 declared on your platform.
262 Don't use "index()" or "rindex()"; instead, use the ANSI C equivalents,
263 "strchr()" and "strrchr()". Not all platforms necessarily have
264 "index()" or "rindex()", and those that do might not declare them in the
265 header file on which they're declared on your platform.
267 Don't fetch data from packets by getting a pointer to data in the packet
268 with "tvb_get_ptr()", casting that pointer to a pointer to a structure,
269 and dereferencing that pointer. That pointer won't necessarily be aligned
270 on the proper boundary, which can cause crashes on some platforms (even
271 if it doesn't crash on an x86-based PC); furthermore, the data in a
272 packet is not necessarily in the byte order of the machine on which
273 Wireshark is running. Use the tvbuff routines to extract individual
274 items from the packet, or use "proto_tree_add_item()" and let it extract
277 Don't use structures that overlay packet data, or into which you copy
278 packet data; the C programming language does not guarantee any
279 particular alignment of fields within a structure, and even the
280 extensions that try to guarantee that are compiler-specific and not
281 necessarily supported by all compilers used to build Wireshark. Using
282 bitfields in those structures is even worse; the order of bitfields
285 Don't use "ntohs()", "ntohl()", "htons()", or "htonl()"; the header
286 files required to define or declare them differ between platforms, and
287 you might be able to get away with not including the appropriate header
288 file on your platform but that might not work on other platforms.
289 Instead, use "g_ntohs()", "g_ntohl()", "g_htons()", and "g_htonl()";
290 those are declared by <glib.h>, and you'll need to include that anyway,
291 as Wireshark header files that all dissectors must include use stuff from
294 Don't fetch a little-endian value using "tvb_get_ntohs() or
295 "tvb_get_ntohl()" and then using "g_ntohs()", "g_htons()", "g_ntohl()",
296 or "g_htonl()" on the resulting value - the g_ routines in question
297 convert between network byte order (big-endian) and *host* byte order,
298 not *little-endian* byte order; not all machines on which Wireshark runs
299 are little-endian, even though PCs are. Fetch those values using
300 "tvb_get_letohs()" and "tvb_get_letohl()".
302 Don't put a comma after the last element of an enum - some compilers may
303 either warn about it (producing extra noise) or refuse to accept it.
305 Don't include <unistd.h> without protecting it with
313 and, if you're including it to get routines such as "open()", "close()",
314 "read()", and "write()" declared, also include <io.h> if present:
320 in order to declare the Windows C library routines "_open()",
321 "_close()", "_read()", and "_write()". Your file must include <glib.h>
322 - which many of the Wireshark header files include, so you might not have
323 to include it explicitly - in order to get "open()", "close()",
324 "read()", "write()", etc. mapped to "_open()", "_close()", "_read()",
327 Do not use "open()", "rename()", "mkdir()", "stat()", "unlink()", "remove()",
328 "fopen()", "freopen()" directly. Instead use "ws_open()", "ws_rename()",
329 "ws_mkdir()", "ws_stat()", "ws_unlink()", "ws_remove()", "ws_fopen()",
330 "ws_freopen()": these wrapper functions change the path and file name from
331 UTF8 to UTF16 on Windows allowing the functions to work correctly when the
332 path or file name contain non-ASCII characters.
334 When opening a file with "ws_fopen()", "ws_freopen()", or "ws_fdopen()", if
335 the file contains ASCII text, use "r", "w", "a", and so on as the open mode
336 - but if it contains binary data, use "rb", "wb", and so on. On
337 Windows, if a file is opened in a text mode, writing a byte with the
338 value of octal 12 (newline) to the file causes two bytes, one with the
339 value octal 15 (carriage return) and one with the value octal 12, to be
340 written to the file, and causes bytes with the value octal 15 to be
341 discarded when reading the file (to translate between C's UNIX-style
342 lines that end with newline and Windows' DEC-style lines that end with
343 carriage return/line feed).
345 In addition, that also means that when opening or creating a binary
346 file, you must use "ws_open()" (with O_CREAT and possibly O_TRUNC if the
347 file is to be created if it doesn't exist), and OR in the O_BINARY flag.
348 That flag is not present on most, if not all, UNIX systems, so you must
355 to properly define it for UNIX (it's not necessary on UNIX).
357 Don't use forward declarations of static arrays without a specified size
358 in a fashion such as this:
360 static const value_string foo_vals[];
364 static const value_string foo_vals[] = {
371 as some compilers will reject the first of those statements. Instead,
372 initialize the array at the point at which it's first declared, so that
375 Don't put a comma after the last tuple of an initializer of an array.
377 For #define names and enum member names, prefix the names with a tag so
378 as to avoid collisions with other names - this might be more of an issue
379 on Windows, as it appears to #define names such as DELETE and
382 Don't use the "numbered argument" feature that many UNIX printf's
385 g_snprintf(add_string, 30, " - (%1$d) (0x%1$04x)", value);
387 as not all UNIX printf's implement it, and Windows printf doesn't appear
388 to implement it. Use something like
390 g_snprintf(add_string, 30, " - (%d) (0x%04x)", value, value);
394 Don't use "variadic macros", such as
396 #define DBG(format, args...) fprintf(stderr, format, ## args)
398 as not all C compilers support them. Use macros that take a fixed
399 number of arguments, such as
401 #define DBG0(format) fprintf(stderr, format)
402 #define DBG1(format, arg1) fprintf(stderr, format, arg1)
403 #define DBG2(format, arg1, arg2) fprintf(stderr, format, arg1, arg2)
409 #define DBG(args) printf args
415 as that's not supported by all compilers.
417 snprintf() -> g_snprintf()
418 snprintf() is not available on all platforms, so it's a good idea to use the
419 g_snprintf() function declared by <glib.h> instead.
421 tmpnam() -> mkstemp()
422 tmpnam is insecure and should not be used any more. Wireshark brings its
423 own mkstemp implementation for use on platforms that lack mkstemp.
424 Note: mkstemp does not accept NULL as a parameter.
426 The pointer returned by a call to "tvb_get_ptr()" is not guaranteed to be
427 aligned on any particular byte boundary; this means that you cannot
428 safely cast it to any data type other than a pointer to "char",
429 "unsigned char", "guint8", or other one-byte data types. You cannot,
430 for example, safely cast it to a pointer to a structure, and then access
431 the structure members directly; on some systems, unaligned accesses to
432 integral data types larger than 1 byte, and floating-point data types,
433 cause a trap, which will, at best, result in the OS slowly performing an
434 unaligned access for you, and will, on at least some platforms, cause
435 the program to be terminated.
437 Wireshark supports platforms with GLib 2.14[.x]/GTK+ 2.12[.x] or newer.
438 If a Glib/GTK+ mechanism is available only in Glib/GTK+ versions newer
439 than 2.14/2.12 then use "#if GLIB_CHECK_VERSION(...)" or "#if
440 GTK_CHECK_VERSION(...)" to conditionally compile code using that
443 When different code must be used on UN*X and Win32, use a #if or #ifdef
444 that tests _WIN32, not WIN32. Try to write code portably whenever
445 possible, however; note that there are some routines in Wireshark with
446 platform-dependent implementations and platform-independent APIs, such
447 as the routines in epan/filesystem.c, allowing the code that calls it to
448 be written portably without #ifdefs.
450 1.1.2 String handling
452 Do not use functions such as strcat() or strcpy().
453 A lot of work has been done to remove the existing calls to these functions and
454 we do not want any new callers of these functions.
456 Instead use g_snprintf() since that function will if used correctly prevent
457 buffer overflows for large strings.
459 When using a buffer to create a string, do not use a buffer stored on the stack.
460 I.e. do not use a buffer declared as
464 instead allocate a buffer dynamically using the string-specific or plain emem
465 routines (see README.malloc) such as
467 emem_strbuf_t *strbuf;
468 strbuf = ep_strbuf_new_label("");
469 ep_strbuf_append_printf(strbuf, ...
475 #define MAX_BUFFER 1024
476 buffer=ep_alloc(MAX_BUFFER);
479 g_snprintf(buffer, MAX_BUFFER, ...
481 This avoids the stack from being corrupted in case there is a bug in your code
482 that accidentally writes beyond the end of the buffer.
485 If you write a routine that will create and return a pointer to a filled in
486 string and if that buffer will not be further processed or appended to after
487 the routine returns (except being added to the proto tree),
488 do not preallocate the buffer to fill in and pass as a parameter instead
489 pass a pointer to a pointer to the function and return a pointer to an
490 emem allocated buffer that will be automatically freed. (see README.malloc)
492 I.e. do not write code such as
494 foo_to_str(char *string, ... ){
500 foo_to_str(buffer, ...
501 proto_tree_add_text(... buffer ...
503 instead write the code as
505 foo_to_str(char **buffer, ...
507 *buffer=ep_alloc(MAX_BUFFER);
513 foo_to_str(&buffer, ...
514 proto_tree_add_text(... *buffer ...
516 Use ep_ allocated buffers. They are very fast and nice. These buffers are all
517 automatically free()d when the dissection of the current packet ends so you
518 don't have to worry about free()ing them explicitly in order to not leak memory.
519 Please read README.malloc.
521 Don't use non-ASCII characters in source files; not all compiler
522 environments will be using the same encoding for non-ASCII characters,
523 and at least one compiler (Microsoft's Visual C) will, in environments
524 with double-byte character encodings, such as many Asian environments,
525 fail if it sees a byte sequence in a source file that doesn't correspond
526 to a valid character. This causes source files using either an ISO
527 8859/n single-byte character encoding or UTF-8 to fail to compile. Even
528 if the compiler doesn't fail, there is no guarantee that the compiler,
529 or a developer's text editor, will interpret the characters the way you
530 intend them to be interpreted.
534 Wireshark is not guaranteed to read only network traces that contain correctly-
535 formed packets. Wireshark is commonly used to track down networking
536 problems, and the problems might be due to a buggy protocol implementation
537 sending out bad packets.
539 Therefore, protocol dissectors not only have to be able to handle
540 correctly-formed packets without, for example, crashing or looping
541 infinitely, they also have to be able to handle *incorrectly*-formed
542 packets without crashing or looping infinitely.
544 Here are some suggestions for making dissectors more robust in the face
545 of incorrectly-formed packets:
547 Do *NOT* use "g_assert()" or "g_assert_not_reached()" in dissectors.
548 *NO* value in a packet's data should be considered "wrong" in the sense
549 that it's a problem with the dissector if found; if it cannot do
550 anything else with a particular value from a packet's data, the
551 dissector should put into the protocol tree an indication that the
552 value is invalid, and should return. The "expert" mechanism should be
553 used for that purpose.
555 If there is a case where you are checking not for an invalid data item
556 in the packet, but for a bug in the dissector (for example, an
557 assumption being made at a particular point in the code about the
558 internal state of the dissector), use the DISSECTOR_ASSERT macro for
559 that purpose; this will put into the protocol tree an indication that
560 the dissector has a bug in it, and will not crash the application.
562 If you are allocating a chunk of memory to contain data from a packet,
563 or to contain information derived from data in a packet, and the size of
564 the chunk of memory is derived from a size field in the packet, make
565 sure all the data is present in the packet before allocating the buffer.
568 1) Wireshark won't leak that chunk of memory if an attempt to
569 fetch data not present in the packet throws an exception.
573 2) it won't crash trying to allocate an absurdly-large chunk of
574 memory if the size field has a bogus large value.
576 If you're fetching into such a chunk of memory a string from the buffer,
577 and the string has a specified size, you can use "tvb_get_*_string()",
578 which will check whether the entire string is present before allocating
579 a buffer for the string, and will also put a trailing '\0' at the end of
582 If you're fetching into such a chunk of memory a 2-byte Unicode string
583 from the buffer, and the string has a specified size, you can use
584 "tvb_get_ephemeral_faked_unicode()", which will check whether the entire
585 string is present before allocating a buffer for the string, and will also
586 put a trailing '\0' at the end of the buffer. The resulting string will be
587 a sequence of single-byte characters; the only Unicode characters that
588 will be handled correctly are those in the ASCII range. (Wireshark's
589 ability to handle non-ASCII strings is limited; it needs to be
592 If you're fetching into such a chunk of memory a sequence of bytes from
593 the buffer, and the sequence has a specified size, you can use
594 "tvb_memdup()", which will check whether the entire sequence is present
595 before allocating a buffer for it.
597 Otherwise, you can check whether the data is present by using
598 "tvb_ensure_bytes_exist()" or by getting a pointer to the data by using
599 "tvb_get_ptr()", although note that there might be problems with using
600 the pointer from "tvb_get_ptr()" (see the item on this in the
601 Portability section above, and the next item below).
603 Note also that you should only fetch string data into a fixed-length
604 buffer if the code ensures that no more bytes than will fit into the
605 buffer are fetched ("the protocol ensures" isn't good enough, as
606 protocol specifications can't ensure only packets that conform to the
607 specification will be transmitted or that only packets for the protocol
608 in question will be interpreted as packets for that protocol by
609 Wireshark). If there's no maximum length of string data to be fetched,
610 routines such as "tvb_get_*_string()" are safer, as they allocate a buffer
611 large enough to hold the string. (Note that some variants of this call
612 require you to free the string once you're finished with it.)
614 If you have gotten a pointer using "tvb_get_ptr()", you must make sure
615 that you do not refer to any data past the length passed as the last
616 argument to "tvb_get_ptr()"; while the various "tvb_get" routines
617 perform bounds checking and throw an exception if you refer to data not
618 available in the tvbuff, direct references through a pointer gotten from
619 "tvb_get_ptr()" do not do any bounds checking.
621 If you have a loop that dissects a sequence of items, each of which has
622 a length field, with the offset in the tvbuff advanced by the length of
623 the item, then, if the length field is the total length of the item, and
624 thus can be zero, you *MUST* check for a zero-length item and abort the
625 loop if you see one. Otherwise, a zero-length item could cause the
626 dissector to loop infinitely. You should also check that the offset,
627 after having the length added to it, is greater than the offset before
628 the length was added to it, if the length field is greater than 24 bits
629 long, so that, if the length value is *very* large and adding it to the
630 offset causes an overflow, that overflow is detected.
634 for (i = {start}; i < {end}; i++)
636 loop, make sure that the type of the loop index variable is large enough
637 to hold the maximum {end} value plus 1; otherwise, the loop index
638 variable can overflow before it ever reaches its maximum value. In
639 particular, be very careful when using gint8, guint8, gint16, or guint16
640 variables as loop indices; you almost always want to use an "int"/"gint"
641 or "unsigned int"/"guint" as the loop index rather than a shorter type.
643 If you are fetching a length field from the buffer, corresponding to the
644 length of a portion of the packet, and subtracting from that length a
645 value corresponding to the length of, for example, a header in the
646 packet portion in question, *ALWAYS* check that the value of the length
647 field is greater than or equal to the length you're subtracting from it,
648 and report an error in the packet and stop dissecting the packet if it's
649 less than the length you're subtracting from it. Otherwise, the
650 resulting length value will be negative, which will either cause errors
651 in the dissector or routines called by the dissector, or, if the value
652 is interpreted as an unsigned integer, will cause the value to be
653 interpreted as a very large positive value.
655 Any tvbuff offset that is added to as processing is done on a packet
656 should be stored in a 32-bit variable, such as an "int"; if you store it
657 in an 8-bit or 16-bit variable, you run the risk of the variable
660 sprintf() -> g_snprintf()
661 Prevent yourself from using the sprintf() function, as it does not test the
662 length of the given output buffer and might be writing into unintended memory
663 areas. This function is one of the main causes of security problems like buffer
664 exploits and many other bugs that are very hard to find. It's much better to
665 use the g_snprintf() function declared by <glib.h> instead.
667 You should test your dissector against incorrectly-formed packets. This
668 can be done using the randpkt and editcap utilities that come with the
669 Wireshark distribution. Testing using randpkt can be done by generating
670 output at the same layer as your protocol, and forcing Wireshark/TShark
671 to decode it as your protocol, e.g. if your protocol sits on top of UDP:
673 randpkt -c 50000 -t dns randpkt.pcap
674 tshark -nVr randpkt.pcap -d udp.port==53,<myproto>
676 Testing using editcap can be done using preexisting capture files and the
677 "-E" flag, which introduces errors in a capture file. E.g.:
679 editcap -E 0.03 infile.pcap outfile.pcap
680 tshark -nVr outfile.pcap
682 The script fuzz-test.sh is available to help automate these tests.
684 1.1.4 Name convention.
686 Wireshark uses the underscore_convention rather than the InterCapConvention for
687 function names, so new code should probably use underscores rather than
688 intercaps for functions and variable names. This is especially important if you
689 are writing code that will be called from outside your code. We are just
690 trying to keep things consistent for other developers.
692 1.1.5 White space convention.
694 Avoid using tab expansions different from 8 column widths, as not all
695 text editors in use by the developers support this. For a detailed
696 discussion of tabs, spaces, and indentation, see
698 http://www.jwz.org/doc/tabs-vs-spaces.html
700 When creating a new file, you are free to choose an indentation logic.
701 Most of the files in Wireshark tend to use 2-space or 4-space
702 indentation. You are encouraged to write a short comment on the
703 indentation logic at the beginning of this new file, especially if
704 you're using non-mod-8 tabs. The tabs-vs-spaces document above provides
705 examples of Emacs and vi modelines for this purpose.
707 Please do not leave trailing whitespace (spaces/tabs) on lines.
709 When editing an existing file, try following the existing indentation
710 logic and even if it very tempting, never ever use a restyler/reindenter
711 utility on an existing file. If you run across wildly varying
712 indentation styles within the same file, it might be helpful to send a
713 note to wireshark-dev for guidance.
715 1.1.6 Compiler warnings
717 You should write code that is free of compiler warnings. Such warnings will
718 often indicate questionable code and sometimes even real bugs, so it's best
719 to avoid warnings at all.
721 The compiler flags in the Makefiles are set to "treat warnings as errors",
722 so your code won't even compile when warnings occur.
726 Wireshark requires certain things when setting up a protocol dissector.
727 Below is skeleton code for a dissector that you can copy to a file and
728 fill in. Your dissector should follow the naming convention of packet-
729 followed by the abbreviated name for the protocol. It is recommended
730 that where possible you keep to the IANA abbreviated name for the
731 protocol, if there is one, or a commonly-used abbreviation for the
734 Usually, you will put your newly created dissector file into the directory
735 epan/dissectors, just like all the other packet-....c files already in there.
737 Also, please add your dissector file to the corresponding makefiles,
738 described in section "1.9 Editing Makefile.common and CMakeLists.txt
739 to add your dissector" below.
741 Dissectors that use the dissector registration to register with a lower level
742 dissector don't need to define a prototype in the .h file. For other
743 dissectors the main dissector routine should have a prototype in a header
744 file whose name is "packet-", followed by the abbreviated name for the
745 protocol, followed by ".h"; any dissector file that calls your dissector
746 should be changed to include that file.
748 You may not need to include all the headers listed in the skeleton
749 below, and you may need to include additional headers.
751 The stdio.h, stdlib.h and string.h header files should be included only as needed.
754 The "$Id$" in the comment will be updated by Subversion when the file is
757 When creating a new file, it is fine to just write "$Id$" as Subversion will
758 automatically fill in the identifier at the time the file will be added to the
759 SVN repository (committed).
761 ------------------------------------Cut here------------------------------------
762 /* packet-PROTOABBREV.c
763 * Routines for PROTONAME dissection
764 * Copyright 201x, YOUR_NAME <YOUR_EMAIL_ADDRESS>
768 * Wireshark - Network traffic analyzer
769 * By Gerald Combs <gerald@wireshark.org>
770 * Copyright 1998 Gerald Combs
772 * Copied from WHATEVER_FILE_YOU_USED (where "WHATEVER_FILE_YOU_USED"
773 * is a dissector file; if you just copied this from README.developer,
774 * don't bother with the "Copied from" - you don't even need to put
775 * in a "Copied from" if you copied an existing dissector, especially
776 * if the bulk of the code in the new dissector is your code)
778 * This program is free software; you can redistribute it and/or modify
779 * it under the terms of the GNU General Public License as published by
780 * the Free Software Foundation; either version 2 of the License, or
781 * (at your option) any later version.
783 * This program is distributed in the hope that it will be useful,
784 * but WITHOUT ANY WARRANTY; without even the implied warranty of
785 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
786 * GNU General Public License for more details.
788 * You should have received a copy of the GNU General Public License along
789 * with this program; if not, write to the Free Software Foundation, Inc.,
790 * 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA.
798 /* Include only as needed */
806 #include <epan/packet.h>
807 #include <epan/prefs.h>
809 /* IF PROTO exposes code to other dissectors, then it must be exported
810 in a header file. If not, a header file is not needed at all. */
811 #include "packet-PROTOABBREV.h"
813 /* Forward declaration we need below (if using proto_reg_handoff...
814 as a prefs callback) */
815 void proto_reg_handoff_PROTOABBREV(void);
817 /* Initialize the protocol and registered fields */
818 static int proto_PROTOABBREV = -1;
819 static int hf_PROTOABBREV_FIELDABBREV = -1;
821 /* Global sample preference ("controls" display of numbers) */
822 static gboolean gPREF_HEX = FALSE;
823 /* Global sample port pref */
824 static guint gPORT_PREF = 1234;
826 /* Initialize the subtree pointers */
827 static gint ett_PROTOABBREV = -1;
829 /* Code to actually dissect the packets */
831 dissect_PROTOABBREV(tvbuff_t *tvb, packet_info *pinfo, proto_tree *tree)
834 /* Set up structures needed to add the protocol subtree and manage it */
836 proto_tree *PROTOABBREV_tree;
838 /* First, if at all possible, do some heuristics to check if the packet cannot
839 * possibly belong to your protocol. This is especially important for
840 * protocols directly on top of TCP or UDP where port collisions are
841 * common place (e.g., even though your protocol uses a well known port,
842 * someone else may set up, for example, a web server on that port which,
843 * if someone analyzed that web server's traffic in Wireshark, would result
844 * in Wireshark handing an HTTP packet to your dissector). For example:
846 /* Check that there's enough data */
847 if (tvb_length(tvb) < /* your protocol's smallest packet size */)
850 /* Get some values from the packet header, probably using tvb_get_*() */
851 if ( /* these values are not possible in PROTONAME */ )
852 /* This packet does not appear to belong to PROTONAME.
853 * Return 0 to give another dissector a chance to dissect it.
857 /* Make entries in Protocol column and Info column on summary display */
858 col_set_str(pinfo->cinfo, COL_PROTOCOL, "PROTOABBREV");
860 /* This field shows up as the "Info" column in the display; you should use
861 it, if possible, to summarize what's in the packet, so that a user looking
862 at the list of packets can tell what type of packet it is. See section 1.5
863 for more information.
865 If you are setting the column to a constant string, use "col_set_str()",
866 as it's more efficient than the other "col_set_XXX()" calls.
868 If you're setting it to a string you've constructed, or will be
869 appending to the column later, use "col_add_str()".
871 "col_add_fstr()" can be used instead of "col_add_str()"; it takes
872 "printf()"-like arguments. Don't use "col_add_fstr()" with a format
873 string of "%s" - just use "col_add_str()" or "col_set_str()", as it's
874 more efficient than "col_add_fstr()".
876 If you will be fetching any data from the packet before filling in
877 the Info column, clear that column first, in case the calls to fetch
878 data from the packet throw an exception because they're fetching data
879 past the end of the packet, so that the Info column doesn't have data
880 left over from the previous dissector; do
882 col_clear(pinfo->cinfo, COL_INFO);
886 col_set_str(pinfo->cinfo, COL_INFO, "XXX Request");
888 /* A protocol dissector may be called in 2 different ways - with, or
889 without a non-null "tree" argument.
891 If the proto_tree argument is null, Wireshark does not need to use
892 the protocol tree information from your dissector, and therefore is
893 passing the dissector a null "tree" argument so that it doesn't
894 need to do work necessary to build the protocol tree.
896 In the interest of speed, if "tree" is NULL, avoid building a
897 protocol tree and adding stuff to it, or even looking at any packet
898 data needed only if you're building the protocol tree, if possible.
900 Note, however, that you must fill in column information, create
901 conversations, reassemble packets, build any other persistent state
902 needed for dissection, and call subdissectors regardless of whether
903 "tree" is NULL or not. This might be inconvenient to do without
904 doing most of the dissection work; the routines for adding items to
905 the protocol tree can be passed a null protocol tree pointer, in
906 which case they'll return a null item pointer, and
907 "proto_item_add_subtree()" returns a null tree pointer if passed a
908 null item pointer, so, if you're careful not to dereference any null
909 tree or item pointers, you can accomplish this by doing all the
910 dissection work. This might not be as efficient as skipping that
911 work if you're not building a protocol tree, but if the code would
912 have a lot of tests whether "tree" is null if you skipped that work,
913 you might still be better off just doing all that work regardless of
914 whether "tree" is null or not.
916 Note also that there is no guarantee, the first time the dissector is
917 called, whether "tree" will be null or not; your dissector must work
918 correctly, building or updating whatever state information is
919 necessary, in either case. */
922 /* NOTE: The offset and length values in the call to
923 "proto_tree_add_item()" define what data bytes to highlight in the hex
924 display window when the line in the protocol tree display
925 corresponding to that item is selected.
927 Supplying a length of -1 is the way to highlight all data from the
928 offset to the end of the packet. */
930 /* create display subtree for the protocol */
931 ti = proto_tree_add_item(tree, proto_PROTOABBREV, tvb, 0, -1, ENC_NA);
933 PROTOABBREV_tree = proto_item_add_subtree(ti, ett_PROTOABBREV);
935 /* add an item to the subtree, see section 1.6 for more information */
936 proto_tree_add_item(PROTOABBREV_tree,
937 hf_PROTOABBREV_FIELDABBREV, tvb, offset, len, ENC_xxx);
940 /* Continue adding tree items to process the packet here */
945 /* If this protocol has a sub-dissector call it here, see section 1.8 */
947 /* Return the amount of data this dissector was able to dissect */
948 return tvb_length(tvb);
952 /* Register the protocol with Wireshark */
954 /* this format is require because a script is used to build the C function
955 that calls all the protocol registration.
959 proto_register_PROTOABBREV(void)
961 module_t *PROTOABBREV_module;
963 /* Setup list of header fields See Section 1.6.1 for details*/
964 static hf_register_info hf[] = {
965 { &hf_PROTOABBREV_FIELDABBREV,
966 { "FIELDNAME", "PROTOABBREV.FIELDABBREV",
967 FIELDTYPE, FIELDDISPLAY, FIELDCONVERT, BITMASK,
968 "FIELDDESCR", HFILL }
972 /* Setup protocol subtree array */
973 static gint *ett[] = {
977 /* Register the protocol name and description */
978 proto_PROTOABBREV = proto_register_protocol("PROTONAME",
979 "PROTOSHORTNAME", "PROTOABBREV");
981 /* Required function calls to register the header fields and subtrees used */
982 proto_register_field_array(proto_PROTOABBREV, hf, array_length(hf));
983 proto_register_subtree_array(ett, array_length(ett));
985 /* Register preferences module (See Section 2.6 for more on preferences) */
986 /* (Registration of a prefs callback is not required if there are no */
987 /* prefs-dependent registration functions (eg: a port pref). */
988 /* See proto_reg_handoff below. */
989 /* If a prefs callback is not needed, use NULL instead of */
990 /* proto_reg_handoff_PROTOABBREV in the following). */
991 PROTOABBREV_module = prefs_register_protocol(proto_PROTOABBREV,
992 proto_reg_handoff_PROTOABBREV);
994 /* Register preferences module under preferences subtree.
995 Use this function instead of prefs_register_protocol if you want to group
996 preferences of several protocols under one preferences subtree.
997 Argument subtree identifies grouping tree node name, several subnodes can be
998 specified usign slash '/' (e.g. "OSI/X.500" - protocol preferences will be
999 accessible under Protocols->OSI->X.500-><PROTOSHORTNAME> preferences node.
1001 PROTOABBREV_module = prefs_register_protocol_subtree(const char *subtree,
1002 proto_PROTOABBREV, proto_reg_handoff_PROTOABBREV);
1004 /* Register a sample preference */
1005 prefs_register_bool_preference(PROTOABBREV_module, "show_hex",
1006 "Display numbers in Hex",
1007 "Enable to display numerical values in hexadecimal.",
1010 /* Register a sample port preference */
1011 prefs_register_uint_preference(PROTOABBREV_module, "tcp.port", "PROTOABBREV TCP Port",
1012 " PROTOABBREV TCP port if other than the default",
1017 /* If this dissector uses sub-dissector registration add a registration routine.
1018 This exact format is required because a script is used to find these
1019 routines and create the code that calls these routines.
1021 If this function is registered as a prefs callback (see prefs_register_protocol
1022 above) this function is also called by preferences whenever "Apply" is pressed;
1023 In that case, it should accommodate being called more than once.
1025 This form of the reg_handoff function is used if if you perform
1026 registration functions which are dependent upon prefs. See below
1027 for a simpler form which can be used if there are no
1028 prefs-dependent registration functions.
1031 proto_reg_handoff_PROTOABBREV(void)
1033 static gboolean initialized = FALSE;
1034 static dissector_handle_t PROTOABBREV_handle;
1035 static int currentPort;
1039 /* Use new_create_dissector_handle() to indicate that dissect_PROTOABBREV()
1040 * returns the number of bytes it dissected (or 0 if it thinks the packet
1041 * does not belong to PROTONAME).
1043 PROTOABBREV_handle = new_create_dissector_handle(dissect_PROTOABBREV,
1049 If you perform registration functions which are dependent upon
1050 prefs the you should de-register everything which was associated
1051 with the previous settings and re-register using the new prefs
1052 settings here. In general this means you need to keep track of
1053 the PROTOABBREV_handle and the value the preference had at the time
1054 you registered. The PROTOABBREV_handle value and the value of the
1055 preference can be saved using local statics in this
1056 function (proto_reg_handoff).
1059 dissector_delete_uint("tcp.port", currentPort, PROTOABBREV_handle);
1062 currentPort = gPORT_PREF;
1064 dissector_add_uint("tcp.port", currentPort, PROTOABBREV_handle);
1069 /* Simple form of proto_reg_handoff_PROTOABBREV which can be used if there are
1070 no prefs-dependent registration function calls.
1074 proto_reg_handoff_PROTOABBREV(void)
1076 dissector_handle_t PROTOABBREV_handle;
1078 /* Use new_create_dissector_handle() to indicate that dissect_PROTOABBREV()
1079 * returns the number of bytes it dissected (or 0 if it thinks the packet
1080 * does not belong to PROTONAME).
1082 PROTOABBREV_handle = new_create_dissector_handle(dissect_PROTOABBREV,
1084 dissector_add_uint("PARENT_SUBFIELD", ID_VALUE, PROTOABBREV_handle);
1089 ------------------------------------Cut here------------------------------------
1091 1.3 Explanation of needed substitutions in code skeleton.
1093 In the above code block the following strings should be substituted with
1096 YOUR_NAME Your name, of course. You do want credit, don't you?
1097 It's the only payment you will receive....
1098 YOUR_EMAIL_ADDRESS Keep those cards and letters coming.
1099 WHATEVER_FILE_YOU_USED Add this line if you are using another file as a
1101 PROTONAME The name of the protocol; this is displayed in the
1102 top-level protocol tree item for that protocol.
1103 PROTOSHORTNAME An abbreviated name for the protocol; this is displayed
1104 in the "Preferences" dialog box if your dissector has
1105 any preferences, in the dialog box of enabled protocols,
1106 and in the dialog box for filter fields when constructing
1107 a filter expression.
1108 PROTOABBREV A name for the protocol for use in filter expressions;
1109 it shall contain only lower-case letters, digits, and
1111 FIELDNAME The displayed name for the header field.
1112 FIELDABBREV The abbreviated name for the header field. (NO SPACES)
1113 FIELDTYPE FT_NONE, FT_BOOLEAN, FT_UINT8, FT_UINT16, FT_UINT24,
1114 FT_UINT32, FT_UINT64, FT_INT8, FT_INT16, FT_INT24, FT_INT32,
1115 FT_INT64, FT_FLOAT, FT_DOUBLE, FT_ABSOLUTE_TIME,
1116 FT_RELATIVE_TIME, FT_STRING, FT_STRINGZ, FT_EBCDIC, FT_EUI64
1117 FT_UINT_STRING, FT_ETHER, FT_BYTES, FT_UINT_BYTES, FT_IPv4,
1118 FT_IPv6, FT_IPXNET, FT_FRAMENUM, FT_PROTOCOL, FT_GUID, FT_OID
1119 FIELDDISPLAY For FT_UINT{8,16,24,32,64} and FT_INT{8,16,24,32,64):
1121 BASE_DEC, BASE_HEX, BASE_OCT, BASE_DEC_HEX, BASE_HEX_DEC,
1122 or BASE_CUSTOM, possibly ORed with BASE_RANGE_STRING
1124 For FT_ABSOLUTE_TIME:
1126 ABSOLUTE_TIME_LOCAL, ABSOLUTE_TIME_UTC, or
1127 ABSOLUTE_TIME_DOY_UTC
1129 For FT_BOOLEAN if BITMASK is non-zero:
1131 Number of bits in the field containing the FT_BOOLEAN
1134 For all other types:
1137 FIELDCONVERT VALS(x), RVALS(x), TFS(x), NULL
1138 BITMASK Usually 0x0 unless using the TFS(x) field conversion.
1139 FIELDDESCR A brief description of the field, or NULL. [Please do not use ""].
1140 PARENT_SUBFIELD Lower level protocol field used for lookup, i.e. "tcp.port"
1141 ID_VALUE Lower level protocol field value that identifies this protocol
1142 For example the TCP or UDP port number
1144 If, for example, PROTONAME is "Internet Bogosity Discovery Protocol",
1145 PROTOSHORTNAME would be "IBDP", and PROTOABBREV would be "ibdp". Try to
1146 conform with IANA names.
1148 1.4 The dissector and the data it receives.
1153 This is only needed if the dissector doesn't use self-registration to
1154 register itself with the lower level dissector, or if the protocol dissector
1155 wants/needs to expose code to other subdissectors.
1157 The dissector must be declared exactly as follows in the file
1158 packet-PROTOABBREV.h:
1161 dissect_PROTOABBREV(tvbuff_t *tvb, packet_info *pinfo, proto_tree *tree);
1164 1.4.2 Extracting data from packets.
1166 NOTE: See the file /epan/tvbuff.h for more details.
1168 The "tvb" argument to a dissector points to a buffer containing the raw
1169 data to be analyzed by the dissector; for example, for a protocol
1170 running atop UDP, it contains the UDP payload (but not the UDP header,
1171 or any protocol headers above it). A tvbuffer is an opaque data
1172 structure, the internal data structures are hidden and the data must be
1173 accessed via the tvbuffer accessors.
1177 Bit accessors for a maximum of 8-bits, 16-bits 32-bits and 64-bits:
1179 guint8 tvb_get_bits8(tvbuff_t *tvb, gint bit_offset, gint no_of_bits);
1180 guint16 tvb_get_bits16(tvbuff_t *tvb, gint bit_offset, gint no_of_bits,gboolean little_endian);
1181 guint32 tvb_get_bits32(tvbuff_t *tvb, gint bit_offset, gint no_of_bits,gboolean little_endian);
1182 guint64 tvb_get_bits64(tvbuff_t *tvb, gint bit_offset, gint no_of_bits,gboolean little_endian);
1184 Single-byte accessor:
1186 guint8 tvb_get_guint8(tvbuff_t*, gint offset);
1188 Network-to-host-order accessors for 16-bit integers (guint16), 24-bit
1189 integers, 32-bit integers (guint32), and 64-bit integers (guint64):
1191 guint16 tvb_get_ntohs(tvbuff_t*, gint offset);
1192 guint32 tvb_get_ntoh24(tvbuff_t*, gint offset);
1193 guint32 tvb_get_ntohl(tvbuff_t*, gint offset);
1194 guint64 tvb_get_ntoh40(tvbuff_t*, gint offset);
1195 guint64 tvb_get_ntoh48(tvbuff_t*, gint offset);
1196 guint64 tvb_get_ntoh56(tvbuff_t*, gint offset);
1197 guint64 tvb_get_ntoh64(tvbuff_t*, gint offset);
1199 Network-to-host-order accessors for single-precision and
1200 double-precision IEEE floating-point numbers:
1202 gfloat tvb_get_ntohieee_float(tvbuff_t*, gint offset);
1203 gdouble tvb_get_ntohieee_double(tvbuff_t*, gint offset);
1205 Little-Endian-to-host-order accessors for 16-bit integers (guint16),
1206 24-bit integers, 32-bit integers (guint32), and 64-bit integers
1209 guint16 tvb_get_letohs(tvbuff_t*, gint offset);
1210 guint32 tvb_get_letoh24(tvbuff_t*, gint offset);
1211 guint32 tvb_get_letohl(tvbuff_t*, gint offset);
1212 guint64 tvb_get_letoh40(tvbuff_t*, gint offset);
1213 guint64 tvb_get_letoh48(tvbuff_t*, gint offset);
1214 guint64 tvb_get_letoh56(tvbuff_t*, gint offset);
1215 guint64 tvb_get_letoh64(tvbuff_t*, gint offset);
1217 Little-Endian-to-host-order accessors for single-precision and
1218 double-precision IEEE floating-point numbers:
1220 gfloat tvb_get_letohieee_float(tvbuff_t*, gint offset);
1221 gdouble tvb_get_letohieee_double(tvbuff_t*, gint offset);
1223 Accessors for IPv4 and IPv6 addresses:
1225 guint32 tvb_get_ipv4(tvbuff_t*, gint offset);
1226 void tvb_get_ipv6(tvbuff_t*, gint offset, struct e_in6_addr *addr);
1228 NOTE: IPv4 addresses are not to be converted to host byte order before
1229 being passed to "proto_tree_add_ipv4()". You should use "tvb_get_ipv4()"
1230 to fetch them, not "tvb_get_ntohl()" *OR* "tvb_get_letohl()" - don't,
1231 for example, try to use "tvb_get_ntohl()", find that it gives you the
1232 wrong answer on the PC on which you're doing development, and try
1233 "tvb_get_letohl()" instead, as "tvb_get_letohl()" will give the wrong
1234 answer on big-endian machines.
1238 void tvb_get_ntohguid(tvbuff_t *, gint offset, e_guid_t *guid);
1239 void tvb_get_letohguid(tvbuff_t *, gint offset, e_guid_t *guid);
1243 guint8 *tvb_get_string(tvbuff_t*, gint offset, gint length);
1244 gchar *tvb_get_unicode_string(tvbuff_t *tvb, const gint offset, gint length, const guint encoding);
1245 guint8 *tvb_get_ephemeral_string(tvbuff_t*, gint offset, gint length);
1246 gchar *tvb_get_ephemeral_unicode_string(tvbuff_t *tvb, const gint offset, gint length, const guint encoding);
1247 guint8 *tvb_get_seasonal_string(tvbuff_t*, gint offset, gint length);
1249 Returns a null-terminated buffer containing data from the specified
1250 tvbuff, starting at the specified offset, and containing the specified
1251 length worth of characters (the length of the buffer will be length+1,
1252 as it includes a null character to terminate the string).
1254 tvb_get_string() returns a buffer allocated by g_malloc() so you must
1255 g_free() it when you are finished with the string. Failure to g_free() this
1256 buffer will lead to memory leaks.
1258 tvb_get_unicode_string() is a unicode (UTF-16) version of above. This
1259 is intended for reading UTF-16 unicode strings out of a tvbuff and
1260 returning them as a UTF-8 string for use in Wireshark. The offset and
1261 returned length pointer are in bytes, not UTF-16 characters.
1263 tvb_get_ephemeral_string() returns a buffer allocated from a special heap
1264 with a lifetime until the next packet is dissected. You do not need to
1265 free() this buffer, it will happen automatically once the next packet is
1268 tvb_get_ephemeral_unicode_string() is a unicode (UTF-16) version of above.
1269 This is intended for reading UTF-16 unicode strings out of a tvbuff and
1270 returning them as a UTF-8 string for use in Wireshark. The offset and
1271 returned length pointer are in bytes, not UTF-16 characters.
1273 tvb_get_seasonal_string() returns a buffer allocated from a special heap
1274 with a lifetime of the current capture session. You do not need to
1275 free() this buffer, it will happen automatically once the a new capture or
1278 guint8 *tvb_get_stringz(tvbuff_t *tvb, gint offset, gint *lengthp);
1279 const guint8 *tvb_get_const stringz(tvbuff_t *tvb, gint offset, gint *lengthp);
1280 guint8 *tvb_get_ephemeral_stringz(tvbuff_t *tvb, gint offset, gint *lengthp);
1281 gchar *tvb_get_ephemeral_unicode_stringz(tvbuff_t *tvb, const gint offset, gint *lengthp, const guint encoding);
1282 guint8 *tvb_get_seasonal_stringz(tvbuff_t *tvb, gint offset, gint *lengthp);
1284 Returns a null-terminated buffer containing data from the specified tvbuff,
1285 starting at the specified offset, and containing all characters from the
1286 tvbuff up to and including a terminating null character in the tvbuff.
1287 "*lengthp" will be set to the length of the string, including the terminating
1290 tvb_get_stringz() returns a buffer allocated by g_malloc() so you must
1291 g_free() it when you are finished with the string. Failure to g_free() this
1292 buffer will lead to memory leaks.
1294 tvb_get_const_stringz() returns a pointer to the (const) string in the tvbuff.
1295 You do not need to free() this buffer, it will happen automatically once the
1296 next packet is dissected. This function is slightly more efficient than the
1297 others because it does not allocate memory and copy the string.
1299 tvb_get_ephemeral_stringz() returns a buffer allocated from a special heap
1300 with a lifetime until the next packet is dissected. You do not need to
1301 free() this buffer, it will happen automatically once the next packet is
1304 tvb_get_ephemeral_unicode_stringz() is a unicode (UTF-16) version of
1305 above. This is intended for reading UTF-16 unicode strings out of a tvbuff
1306 and returning them as a UTF-8 string for use in Wireshark. The offset and
1307 returned length pointer are in bytes, not UTF-16 characters.
1309 tvb_get_seasonal_stringz() returns a buffer allocated from a special heap
1310 with a lifetime of the current capture session. You do not need to
1311 free() this buffer, it will happen automatically once the a new capture or
1314 tvb_fake_unicode() has been superceded by tvb_get_unicode_string(), which
1315 properly handles Unicode (UTF-16) strings by converting them to UTF-8.
1317 tvb_get_ephemeral_faked_unicode() has been superceded by tvb_get_ephemeral_string(), which properly handles Unicode (UTF-16) strings by converting them
1320 Byte Array Accessors:
1322 gchar *tvb_bytes_to_str(tvbuff_t *tvb, gint offset, gint len);
1324 Formats a bunch of data from a tvbuff as bytes, returning a pointer
1325 to the string with the data formatted as two hex digits for each byte.
1326 The string pointed to is stored in an "ep_alloc'd" buffer which will be freed
1327 before the next frame is dissected. The formatted string will contain the hex digits
1328 for at most the first 16 bytes of the data. If len is greater than 16 bytes, a
1329 trailing "..." will be added to the string.
1331 gchar *tvb_bytes_to_str_punct(tvbuff_t *tvb, gint offset, gint len, gchar punct);
1333 This function is similar to tvb_bytes_to_str(...) except that 'punct' is inserted
1334 between the hex representation of each byte.
1336 gchar *tvb_bcd_dig_to_ep_str(tvbuff_t *tvb, const gint offset, const gint len, dgt_set_t *dgt, gboolean skip_first);
1338 Given a tvbuff, an offset into the tvbuff, and a length that starts
1339 at that offset (which may be -1 for "all the way to the end of the
1340 tvbuff"), fetch BCD encoded digits from a tvbuff starting from either
1341 the low or high half byte, formating the digits according to an input digit set,
1342 if NUll a default digit set of 0-9 returning "?" for overdecadic digits will be used.
1343 A pointer to the EP allocated string will be returned.
1344 Note: a tvbuff content of 0xf is considered a 'filler' and will end the conversion.
1347 guint8* tvb_memcpy(tvbuff_t*, guint8* target, gint offset, gint length);
1349 Copies into the specified target the specified length's worth of data
1350 from the specified tvbuff, starting at the specified offset.
1352 guint8* tvb_memdup(tvbuff_t*, gint offset, gint length);
1353 guint8* ep_tvb_memdup(tvbuff_t*, gint offset, gint length);
1355 Returns a buffer, allocated with "g_malloc()", containing the specified
1356 length's worth of data from the specified tvbuff, starting at the
1357 specified offset. The ephemeral variant is freed automatically after the
1358 packet is dissected.
1361 /* WARNING! This function is possibly expensive, temporarily allocating
1362 * another copy of the packet data. Furthermore, it's dangerous because once
1363 * this pointer is given to the user, there's no guarantee that the user will
1364 * honor the 'length' and not overstep the boundaries of the buffer.
1366 guint8* tvb_get_ptr(tvbuff_t*, gint offset, gint length);
1368 The reason that tvb_get_ptr() might have to allocate a copy of its data
1369 only occurs with TVBUFF_COMPOSITES, data that spans multiple tvbuffers.
1370 If the user requests a pointer to a range of bytes that span the member
1371 tvbuffs that make up the TVBUFF_COMPOSITE, the data will have to be
1372 copied to another memory region to assure that all the bytes are
1377 1.5 Functions to handle columns in the traffic summary window.
1379 The topmost pane of the main window is a list of the packets in the
1380 capture, possibly filtered by a display filter.
1382 Each line corresponds to a packet, and has one or more columns, as
1383 configured by the user.
1385 Many of the columns are handled by code outside individual dissectors;
1386 most dissectors need only specify the value to put in the "Protocol" and
1389 Columns are specified by COL_ values; the COL_ value for the "Protocol"
1390 field, typically giving an abbreviated name for the protocol (but not
1391 the all-lower-case abbreviation used elsewhere) is COL_PROTOCOL, and the
1392 COL_ value for the "Info" field, giving a summary of the contents of the
1393 packet for that protocol, is COL_INFO.
1395 The value for a column can be specified with one of several functions,
1396 all of which take the 'fd' argument to the dissector as their first
1397 argument, and the COL_ value for the column as their second argument.
1399 1.5.1 The col_set_str function.
1401 'col_set_str' takes a string as its third argument, and sets the value
1402 for the column to that value. It assumes that the pointer passed to it
1403 points to a string constant or a static "const" array, not to a
1404 variable, as it doesn't copy the string, it merely saves the pointer
1405 value; the argument can itself be a variable, as long as it always
1406 points to a string constant or a static "const" array.
1408 It is more efficient than 'col_add_str' or 'col_add_fstr'; however, if
1409 the dissector will be using 'col_append_str' or 'col_append_fstr" to
1410 append more information to the column, the string will have to be copied
1411 anyway, so it's best to use 'col_add_str' rather than 'col_set_str' in
1414 For example, to set the "Protocol" column
1417 col_set_str(pinfo->cinfo, COL_PROTOCOL, "PROTOABBREV");
1420 1.5.2 The col_add_str function.
1422 'col_add_str' takes a string as its third argument, and sets the value
1423 for the column to that value. It takes the same arguments as
1424 'col_set_str', but copies the string, so that if the string is, for
1425 example, an automatic variable that won't remain in scope when the
1426 dissector returns, it's safe to use.
1429 1.5.3 The col_add_fstr function.
1431 'col_add_fstr' takes a 'printf'-style format string as its third
1432 argument, and 'printf'-style arguments corresponding to '%' format
1433 items in that string as its subsequent arguments. For example, to set
1434 the "Info" field to "<XXX> request, <N> bytes", where "reqtype" is a
1435 string containing the type of the request in the packet and "n" is an
1436 unsigned integer containing the number of bytes in the request:
1438 col_add_fstr(pinfo->cinfo, COL_INFO, "%s request, %u bytes",
1441 Don't use 'col_add_fstr' with a format argument of just "%s" -
1442 'col_add_str', or possibly even 'col_set_str' if the string that matches
1443 the "%s" is a static constant string, will do the same job more
1447 1.5.4 The col_clear function.
1449 If the Info column will be filled with information from the packet, that
1450 means that some data will be fetched from the packet before the Info
1451 column is filled in. If the packet is so small that the data in
1452 question cannot be fetched, the routines to fetch the data will throw an
1453 exception (see the comment at the beginning about tvbuffers improving
1454 the handling of short packets - the tvbuffers keep track of how much
1455 data is in the packet, and throw an exception on an attempt to fetch
1456 data past the end of the packet, so that the dissector won't process
1457 bogus data), causing the Info column not to be filled in.
1459 This means that the Info column will have data for the previous
1460 protocol, which would be confusing if, for example, the Protocol column
1461 had data for this protocol.
1463 Therefore, before a dissector fetches any data whatsoever from the
1464 packet (unless it's a heuristic dissector fetching data to determine
1465 whether the packet is one that it should dissect, in which case it
1466 should check, before fetching the data, whether there's any data to
1467 fetch; if there isn't, it should return FALSE), it should set the
1468 Protocol column and the Info column.
1470 If the Protocol column will ultimately be set to, for example, a value
1471 containing a protocol version number, with the version number being a
1472 field in the packet, the dissector should, before fetching the version
1473 number field or any other field from the packet, set it to a value
1474 without a version number, using 'col_set_str', and should later set it
1475 to a value with the version number after it's fetched the version
1478 If the Info column will ultimately be set to a value containing
1479 information from the packet, the dissector should, before fetching any
1480 fields from the packet, clear the column using 'col_clear' (which is
1481 more efficient than clearing it by calling 'col_set_str' or
1482 'col_add_str' with a null string), and should later set it to the real
1483 string after it's fetched the data to use when doing that.
1486 1.5.5 The col_append_str function.
1488 Sometimes the value of a column, especially the "Info" column, can't be
1489 conveniently constructed at a single point in the dissection process;
1490 for example, it might contain small bits of information from many of the
1491 fields in the packet. 'col_append_str' takes, as arguments, the same
1492 arguments as 'col_add_str', but the string is appended to the end of the
1493 current value for the column, rather than replacing the value for that
1494 column. (Note that no blank separates the appended string from the
1495 string to which it is appended; if you want a blank there, you must add
1496 it yourself as part of the string being appended.)
1499 1.5.6 The col_append_fstr function.
1501 'col_append_fstr' is to 'col_add_fstr' as 'col_append_str' is to
1502 'col_add_str' - it takes, as arguments, the same arguments as
1503 'col_add_fstr', but the formatted string is appended to the end of the
1504 current value for the column, rather than replacing the value for that
1507 1.5.7 The col_append_sep_str and col_append_sep_fstr functions.
1509 In specific situations the developer knows that a column's value will be
1510 created in a stepwise manner, where the appended values are listed. Both
1511 'col_append_sep_str' and 'col_append_sep_fstr' functions will add an item
1512 separator between two consecutive items, and will not add the separator at the
1513 beginning of the column. The remainder of the work both functions do is
1514 identical to what 'col_append_str' and 'col_append_fstr' do.
1516 1.5.8 The col_set_fence and col_prepend_fence_fstr functions.
1518 Sometimes a dissector may be called multiple times for different PDUs in the
1519 same frame (for example in the case of SCTP chunk bundling: several upper
1520 layer data packets may be contained in one SCTP packet). If the upper layer
1521 dissector calls 'col_set_str()' or 'col_clear()' on the Info column when it
1522 begins dissecting each of those PDUs then when the frame is fully dissected
1523 the Info column would contain only the string from the last PDU in the frame.
1524 The 'col_set_fence' function erects a "fence" in the column that prevents
1525 subsequent 'col_...' calls from clearing the data currently in that column.
1526 For example, the SCTP dissector calls 'col_set_fence' on the Info column
1527 after it has called any subdissectors for that chunk so that subdissectors
1528 of any subsequent chunks may only append to the Info column.
1529 'col_prepend_fence_fstr' prepends data before a fence (moving it if
1530 necessary). It will create a fence at the end of the prepended data if the
1531 fence does not already exist.
1534 1.5.9 The col_set_time function.
1536 The 'col_set_time' function takes an nstime value as its third argument.
1537 This nstime value is a relative value and will be added as such to the
1538 column. The fourth argument is the filtername holding this value. This
1539 way, rightclicking on the column makes it possible to build a filter
1540 based on the time-value.
1544 nstime_delta(&ts, &pinfo->fd->abs_ts, &tcpd->ts_first);
1545 col_set_time(pinfo->cinfo, COL_REL_CONV_TIME, &ts, "tcp.time_relative");
1548 1.6 Constructing the protocol tree.
1550 The middle pane of the main window, and the topmost pane of a packet
1551 popup window, are constructed from the "protocol tree" for a packet.
1553 The protocol tree, or proto_tree, is a GNode, the N-way tree structure
1554 available within GLIB. Of course the protocol dissectors don't care
1555 what a proto_tree really is; they just pass the proto_tree pointer as an
1556 argument to the routines which allow them to add items and new branches
1559 When a packet is selected in the packet-list pane, or a packet popup
1560 window is created, a new logical protocol tree (proto_tree) is created.
1561 The pointer to the proto_tree (in this case, 'protocol tree'), is passed
1562 to the top-level protocol dissector, and then to all subsequent protocol
1563 dissectors for that packet, and then the GUI tree is drawn via
1566 The logical proto_tree needs to know detailed information about the protocols
1567 and fields about which information will be collected from the dissection
1568 routines. By strictly defining (or "typing") the data that can be attached to a
1569 proto tree, searching and filtering becomes possible. This means that for
1570 every protocol and field (which I also call "header fields", since they are
1571 fields in the protocol headers) which might be attached to a tree, some
1572 information is needed.
1574 Every dissector routine will need to register its protocols and fields
1575 with the central protocol routines (in proto.c). At first I thought I
1576 might keep all the protocol and field information about all the
1577 dissectors in one file, but decentralization seemed like a better idea.
1578 That one file would have gotten very large; one small change would have
1579 required a re-compilation of the entire file. Also, by allowing
1580 registration of protocols and fields at run-time, loadable modules of
1581 protocol dissectors (perhaps even user-supplied) is feasible.
1583 To do this, each protocol should have a register routine, which will be
1584 called when Wireshark starts. The code to call the register routines is
1585 generated automatically; to arrange that a protocol's register routine
1586 be called at startup:
1588 the file containing a dissector's "register" routine must be
1589 added to "DISSECTOR_SRC" in "epan/dissectors/Makefile.common"
1590 (and in "epan/CMakeLists.txt");
1592 the "register" routine must have a name of the form
1593 "proto_register_XXX";
1595 the "register" routine must take no argument, and return no
1598 the "register" routine's name must appear in the source file
1599 either at the beginning of the line, or preceded only by "void "
1600 at the beginning of the line (that would typically be the
1601 definition) - other white space shouldn't cause a problem, e.g.:
1603 void proto_register_XXX(void) {
1612 proto_register_XXX( void )
1619 and so on should work.
1621 For every protocol or field that a dissector wants to register, a variable of
1622 type int needs to be used to keep track of the protocol. The IDs are
1623 needed for establishing parent/child relationships between protocols and
1624 fields, as well as associating data with a particular field so that it
1625 can be stored in the logical tree and displayed in the GUI protocol
1628 Some dissectors will need to create branches within their tree to help
1629 organize header fields. These branches should be registered as header
1630 fields. Only true protocols should be registered as protocols. This is
1631 so that a display filter user interface knows how to distinguish
1632 protocols from fields.
1634 A protocol is registered with the name of the protocol and its
1637 Here is how the frame "protocol" is registered.
1641 proto_frame = proto_register_protocol (
1643 /* short name */ "Frame",
1644 /* abbrev */ "frame" );
1646 A header field is also registered with its name and abbreviation, but
1647 information about its data type is needed. It helps to look at
1648 the header_field_info struct to see what information is expected:
1650 struct header_field_info {
1655 const void *strings;
1663 A string representing the name of the field. This is the name
1664 that will appear in the graphical protocol tree. It must be a non-empty
1669 A string with an abbreviation of the field. We concatenate the
1670 abbreviation of the parent protocol with an abbreviation for the field,
1671 using a period as a separator. For example, the "src" field in an IP packet
1672 would have "ip.src" as an abbreviation. It is acceptable to have
1673 multiple levels of periods if, for example, you have fields in your
1674 protocol that are then subdivided into subfields. For example, TRMAC
1675 has multiple error fields, so the abbreviations follow this pattern:
1676 "trmac.errors.iso", "trmac.errors.noniso", etc.
1678 The abbreviation is the identifier used in a display filter. If it is
1679 an empty string then the field will not be filterable.
1683 The type of value this field holds. The current field types are:
1685 FT_NONE No field type. Used for fields that
1686 aren't given a value, and that can only
1687 be tested for presence or absence; a
1688 field that represents a data structure,
1689 with a subtree below it containing
1690 fields for the members of the structure,
1691 or that represents an array with a
1692 subtree below it containing fields for
1693 the members of the array, might be an
1695 FT_PROTOCOL Used for protocols which will be placing
1696 themselves as top-level items in the
1697 "Packet Details" pane of the UI.
1698 FT_BOOLEAN 0 means "false", any other value means
1700 FT_FRAMENUM A frame number; if this is used, the "Go
1701 To Corresponding Frame" menu item can
1703 FT_UINT8 An 8-bit unsigned integer.
1704 FT_UINT16 A 16-bit unsigned integer.
1705 FT_UINT24 A 24-bit unsigned integer.
1706 FT_UINT32 A 32-bit unsigned integer.
1707 FT_UINT64 A 64-bit unsigned integer.
1708 FT_INT8 An 8-bit signed integer.
1709 FT_INT16 A 16-bit signed integer.
1710 FT_INT24 A 24-bit signed integer.
1711 FT_INT32 A 32-bit signed integer.
1712 FT_INT64 A 64-bit signed integer.
1713 FT_FLOAT A single-precision floating point number.
1714 FT_DOUBLE A double-precision floating point number.
1715 FT_ABSOLUTE_TIME An absolute time from some fixed point in time,
1716 displayed as the date, followed by the time, as
1717 hours, minutes, and seconds with 9 digits after
1720 Two absolute time encodings may be specified
1721 with proto_tree_add_item():
1723 ENC_TIME_TIMESPEC: Seconds (4 bytes) and
1724 nanoseconds (4 bytes) of time since January 1,
1727 ENC_TIME_NTP: NTP timestamps are represented as
1728 a 64-bit unsigned fixed-point number, in seconds
1729 relative to 0h on 1 January 1900. The integer
1730 part is in the first 32 bits and the fraction
1731 part in the last 32 bits.
1733 The encoding must be logically ORed with
1734 ENC_BIG_ENDIAN or ENC_LITTLE_ENDIAN as
1736 FT_RELATIVE_TIME Seconds (4 bytes) and nanoseconds (4 bytes)
1737 of time relative to an arbitrary time.
1738 displayed as seconds and 9 digits
1739 after the decimal point.
1740 FT_STRING A string of characters, not necessarily
1741 NUL-terminated, but possibly NUL-padded.
1742 This, and the other string-of-characters
1743 types, are to be used for text strings,
1744 not raw binary data.
1745 FT_STRINGZ A NUL-terminated string of characters.
1746 The string length is normally the length
1747 given in the proto_tree_add_item() call.
1748 However if the length given in the call
1749 is -1, then the length used is that
1750 returned by calling tvb_strsize().
1751 FT_EBCDIC A string of characters, not necessarily
1752 NUL-terminated, but possibly NUL-padded.
1753 The data from the packet is converted from
1754 EBCDIC to ASCII before displaying to the user.
1755 FT_UINT_STRING A counted string of characters, consisting
1756 of a count (represented as an integral value,
1757 of width given in the proto_tree_add_item()
1758 call) followed immediately by that number of
1760 FT_ETHER A six octet string displayed in
1761 Ethernet-address format.
1762 FT_BYTES A string of bytes with arbitrary values;
1763 used for raw binary data.
1764 FT_UINT_BYTES A counted string of bytes, consisting
1765 of a count (represented as an integral value,
1766 of width given in the proto_tree_add_item()
1767 call) followed immediately by that number of
1768 arbitrary values; used for raw binary data.
1769 FT_IPv4 A version 4 IP address (4 bytes) displayed
1770 in dotted-quad IP address format (4
1771 decimal numbers separated by dots).
1772 FT_IPv6 A version 6 IP address (16 bytes) displayed
1773 in standard IPv6 address format.
1774 FT_IPXNET An IPX address displayed in hex as a 6-byte
1775 network number followed by a 6-byte station
1777 FT_GUID A Globally Unique Identifier
1778 FT_OID An ASN.1 Object Identifier
1779 FT_EUI64 A EUI-64 Address
1781 Some of these field types are still not handled in the display filter
1782 routines, but the most common ones are. The FT_UINT* variables all
1783 represent unsigned integers, and the FT_INT* variables all represent
1784 signed integers; the number on the end represent how many bits are used
1785 to represent the number.
1787 Some constraints are imposed on the header fields depending on the type
1788 (e.g. FT_BYTES) of the field. Fields of type FT_ABSOLUTE_TIME must use
1789 'ABSOLUTE_TIME_{LOCAL,UTC,DOY_UTC}, NULL, 0x0' as values for the
1790 'display, 'strings', and 'bitmask' fields, and all other non-integral
1791 types (i.e.. types that are _not_ FT_INT* and FT_UINT*) must use
1792 'BASE_NONE, NULL, 0x0' as values for the 'display', 'strings', 'bitmask'
1793 fields. The reason is simply that the type itself implictly defines the
1794 nature of 'display', 'strings', 'bitmask'.
1798 The display field has a couple of overloaded uses. This is unfortunate,
1799 but since we're using C as an application programming language, this sometimes
1800 makes for cleaner programs. Right now I still think that overloading
1801 this variable was okay.
1803 For integer fields (FT_UINT* and FT_INT*), this variable represents the
1804 base in which you would like the value displayed. The acceptable bases
1814 BASE_DEC, BASE_HEX, and BASE_OCT are decimal, hexadecimal, and octal,
1815 respectively. BASE_DEC_HEX and BASE_HEX_DEC display value in two bases
1816 (the 1st representation followed by the 2nd in parenthesis).
1818 BASE_CUSTOM allows one to specify a callback function pointer that will
1819 format the value. The function pointer of the same type as defined by
1820 custom_fmt_func_t in epan/proto.h, specifically:
1822 void func(gchar *, guint32);
1824 The first argument is a pointer to a buffer of the ITEM_LABEL_LENGTH size
1825 and the second argument is the value to be formatted.
1827 For FT_BOOLEAN fields that are also bitfields (i.e. 'bitmask' is non-zero),
1828 'display' is used to tell the proto_tree how wide the parent bitfield is.
1829 With integers this is not needed since the type of integer itself
1830 (FT_UINT8, FT_UINT16, FT_UINT24, FT_UINT32, etc.) tells the proto_tree how
1831 wide the parent bitfield is.
1833 For FT_ABSOLUTE_TIME fields, 'display' is used to indicate whether the
1834 time is to be displayed as a time in the time zone for the machine on
1835 which Wireshark/TShark is running or as UTC and, for UTC, whether the
1836 date should be displayed as "{monthname}, {month} {day_of_month},
1837 {year}" or as "{year/day_of_year}".
1839 Additionally, BASE_NONE is used for 'display' as a NULL-value. That is, for
1840 non-integers other than FT_ABSOLUTE_TIME fields, and non-bitfield
1841 FT_BOOLEANs, you'll want to use BASE_NONE in the 'display' field. You may
1842 not use BASE_NONE for integers.
1844 It is possible that in the future we will record the endianness of
1845 integers. If so, it is likely that we'll use a bitmask on the display field
1846 so that integers would be represented as BEND|BASE_DEC or LEND|BASE_HEX.
1847 But that has not happened yet; note that there are protocols for which
1848 no endianness is specified, such as the X11 protocol and the DCE RPC
1849 protocol, so it would not be possible to record the endianness of all
1855 Some integer fields, of type FT_UINT*, need labels to represent the true
1856 value of a field. You could think of those fields as having an
1857 enumerated data type, rather than an integral data type.
1859 A 'value_string' structure is a way to map values to strings.
1861 typedef struct _value_string {
1866 For fields of that type, you would declare an array of "value_string"s:
1868 static const value_string valstringname[] = {
1869 { INTVAL1, "Descriptive String 1" },
1870 { INTVAL2, "Descriptive String 2" },
1874 (the last entry in the array must have a NULL 'strptr' value, to
1875 indicate the end of the array). The 'strings' field would be set to
1876 'VALS(valstringname)'.
1878 If the field has a numeric rather than an enumerated type, the 'strings'
1879 field would be set to NULL.
1881 -- Extended value strings
1882 You can also use an extended version of the value_string for faster lookups.
1883 It requires a value_string as input.
1884 If all of a contiguous range of values from min to max are present in the array
1885 the value will be used as as a direct index into a value_string array.
1887 If the values in the array are not contiguous (ie: there are "gaps"), but are in assending order
1888 a binary search will be used.
1890 Note: "gaps" in a value_string array can be filled with "empty" entries eg: {value, "Unknown"} so that
1891 direct access to the array is is possible.
1893 The init macro (see below) will perform a check on the value string
1894 the first time it is used to determine which search algorithm fits and fall back to a linear search
1895 if the value_string does not meet the criteria above.
1897 Use this macro to initialise the extended value_string at comile time:
1899 static value_string_ext valstringname_ext = VALUE_STRING_EXT_INIT(valstringname);
1901 Extended value strings can be created at runtime by calling
1902 value_string_ext_new(<ptr to value_string array>,
1903 <total number of entries in the value_string_array>, /* include {0, NULL} entry */
1904 <value_string_name>);
1906 For hf[] array FT_(U)INT* fields that need a 'valstringname_ext' struct, the 'strings' field
1907 would be set to '&valstringname_ext)'. Furthermore, 'display' field must be
1908 ORed with 'BASE_EXT_STRING' (e.g. BASE_DEC|BASE_EXT_STRING).
1912 If the field has a numeric type that might logically fit in ranges of values
1913 one can use a range_string struct.
1915 Thus a 'range_string' structure is a way to map ranges to strings.
1917 typedef struct _range_string {
1920 const gchar *strptr;
1923 For fields of that type, you would declare an array of "range_string"s:
1925 static const range_string rvalstringname[] = {
1926 { INTVAL_MIN1, INTVALMAX1, "Descriptive String 1" },
1927 { INTVAL_MIN2, INTVALMAX2, "Descriptive String 2" },
1931 If INTVAL_MIN equals INTVAL_MAX for a given entry the range_string
1932 behavior collapses to the one of value_string.
1933 For FT_(U)INT* fields that need a 'range_string' struct, the 'strings' field
1934 would be set to 'RVALS(rvalstringname)'. Furthermore, 'display' field must be
1935 ORed with 'BASE_RANGE_STRING' (e.g. BASE_DEC|BASE_RANGE_STRING).
1938 FT_BOOLEANs have a default map of 0 = "False", 1 (or anything else) = "True".
1939 Sometimes it is useful to change the labels for boolean values (e.g.,
1940 to "Yes"/"No", "Fast"/"Slow", etc.). For these mappings, a struct called
1941 true_false_string is used.
1943 typedef struct true_false_string {
1946 } true_false_string;
1948 For Boolean fields for which "False" and "True" aren't the desired
1949 labels, you would declare a "true_false_string"s:
1951 static const true_false_string boolstringname = {
1956 Its two fields are pointers to the string representing truth, and the
1957 string representing falsehood. For FT_BOOLEAN fields that need a
1958 'true_false_string' struct, the 'strings' field would be set to
1959 'TFS(&boolstringname)'.
1961 If the Boolean field is to be displayed as "False" or "True", the
1962 'strings' field would be set to NULL.
1964 Wireshark predefines a whole range of ready made "true_false_string"s
1965 in tfs.h, included via packet.h.
1969 If the field is a bitfield, then the bitmask is the mask which will
1970 leave only the bits needed to make the field when ANDed with a value.
1971 The proto_tree routines will calculate 'bitshift' automatically
1972 from 'bitmask', by finding the rightmost set bit in the bitmask.
1973 This shift is applied before applying string mapping functions or
1975 If the field is not a bitfield, then bitmask should be set to 0.
1979 This is a string giving a proper description of the field. It should be
1980 at least one grammatically complete sentence, or NULL in which case the
1981 name field is used. (Please do not use "").
1982 It is meant to provide a more detailed description of the field than the
1983 name alone provides. This information will be used in the man page, and
1984 in a future GUI display-filter creation tool. We might also add tooltips
1985 to the labels in the GUI protocol tree, in which case the blurb would
1986 be used as the tooltip text.
1989 1.6.1 Field Registration.
1991 Protocol registration is handled by creating an instance of the
1992 header_field_info struct (or an array of such structs), and
1993 calling the registration function along with the registration ID of
1994 the protocol that is the parent of the fields. Here is a complete example:
1996 static int proto_eg = -1;
1997 static int hf_field_a = -1;
1998 static int hf_field_b = -1;
2000 static hf_register_info hf[] = {
2003 { "Field A", "proto.field_a", FT_UINT8, BASE_HEX, NULL,
2004 0xf0, "Field A represents Apples", HFILL }},
2007 { "Field B", "proto.field_b", FT_UINT16, BASE_DEC, VALS(vs),
2008 0x0, "Field B represents Bananas", HFILL }}
2011 proto_eg = proto_register_protocol("Example Protocol",
2013 proto_register_field_array(proto_eg, hf, array_length(hf));
2015 Be sure that your array of hf_register_info structs is declared 'static',
2016 since the proto_register_field_array() function does not create a copy
2017 of the information in the array... it uses that static copy of the
2018 information that the compiler created inside your array. Here's the
2019 layout of the hf_register_info struct:
2021 typedef struct hf_register_info {
2022 int *p_id; /* pointer to parent variable */
2023 header_field_info hfinfo;
2026 Also be sure to use the handy array_length() macro found in packet.h
2027 to have the compiler compute the array length for you at compile time.
2029 If you don't have any fields to register, do *NOT* create a zero-length
2030 "hf" array; not all compilers used to compile Wireshark support them.
2031 Just omit the "hf" array, and the "proto_register_field_array()" call,
2034 It is OK to have header fields with a different format be registered with
2035 the same abbreviation. For instance, the following is valid:
2037 static hf_register_info hf[] = {
2039 { &hf_field_8bit, /* 8-bit version of proto.field */
2040 { "Field (8 bit)", "proto.field", FT_UINT8, BASE_DEC, NULL,
2041 0x00, "Field represents FOO", HFILL }},
2043 { &hf_field_32bit, /* 32-bit version of proto.field */
2044 { "Field (32 bit)", "proto.field", FT_UINT32, BASE_DEC, NULL,
2045 0x00, "Field represents FOO", HFILL }}
2048 This way a filter expression can match a header field, irrespective of the
2049 representation of it in the specific protocol context. This is interesting
2050 for protocols with variable-width header fields.
2052 The HFILL macro at the end of the struct will set reasonable default values
2053 for internally used fields.
2055 1.6.2 Adding Items and Values to the Protocol Tree.
2057 A protocol item is added to an existing protocol tree with one of a
2058 handful of proto_XXX_DO_YYY() functions.
2060 Remember that it only makes sense to add items to a protocol tree if its
2061 proto_tree pointer is not null. Should you add an item to a NULL tree, then
2062 the proto_XXX_DO_YYY() function will immediately return. The cost of this
2063 function call can be avoided by checking for the tree pointer.
2065 Subtrees can be made with the proto_item_add_subtree() function:
2067 item = proto_tree_add_item(....);
2068 new_tree = proto_item_add_subtree(item, tree_type);
2070 This will add a subtree under the item in question; a subtree can be
2071 created under an item made by any of the "proto_tree_add_XXX" functions,
2072 so that the tree can be given an arbitrary depth.
2074 Subtree types are integers, assigned by
2075 "proto_register_subtree_array()". To register subtree types, pass an
2076 array of pointers to "gint" variables to hold the subtree type values to
2077 "proto_register_subtree_array()":
2079 static gint ett_eg = -1;
2080 static gint ett_field_a = -1;
2082 static gint *ett[] = {
2087 proto_register_subtree_array(ett, array_length(ett));
2089 in your "register" routine, just as you register the protocol and the
2090 fields for that protocol.
2092 The ett_ variables identify particular type of subtree so that if you expand
2093 one of them, Wireshark keeps track of that and, when you click on
2094 another packet, it automatically opens all subtrees of that type.
2095 If you close one of them, all subtrees of that type will be closed when
2096 you move to another packet.
2098 There are several functions that the programmer can use to add either
2099 protocol or field labels to the proto_tree:
2102 proto_tree_add_item(tree, id, tvb, start, length, encoding);
2105 proto_tree_add_none_format(tree, id, tvb, start, length, format, ...);
2108 proto_tree_add_protocol_format(tree, id, tvb, start, length,
2112 proto_tree_add_bytes(tree, id, tvb, start, length, start_ptr);
2115 proto_tree_add_bytes_format(tree, id, tvb, start, length, start_ptr,
2119 proto_tree_add_bytes_format_value(tree, id, tvb, start, length,
2120 start_ptr, format, ...);
2123 proto_tree_add_time(tree, id, tvb, start, length, value_ptr);
2126 proto_tree_add_time_format(tree, id, tvb, start, length, value_ptr,
2130 proto_tree_add_time_format_value(tree, id, tvb, start, length,
2131 value_ptr, format, ...);
2134 proto_tree_add_ipxnet(tree, id, tvb, start, length, value);
2137 proto_tree_add_ipxnet_format(tree, id, tvb, start, length, value,
2141 proto_tree_add_ipxnet_format_value(tree, id, tvb, start, length,
2142 value, format, ...);
2145 proto_tree_add_ipv4(tree, id, tvb, start, length, value);
2148 proto_tree_add_ipv4_format(tree, id, tvb, start, length, value,
2152 proto_tree_add_ipv4_format_value(tree, id, tvb, start, length,
2153 value, format, ...);
2156 proto_tree_add_ipv6(tree, id, tvb, start, length, value_ptr);
2159 proto_tree_add_ipv6_format(tree, id, tvb, start, length, value_ptr,
2163 proto_tree_add_ipv6_format_value(tree, id, tvb, start, length,
2164 value_ptr, format, ...);
2167 proto_tree_add_ether(tree, id, tvb, start, length, value_ptr);
2170 proto_tree_add_ether_format(tree, id, tvb, start, length, value_ptr,
2174 proto_tree_add_ether_format_value(tree, id, tvb, start, length,
2175 value_ptr, format, ...);
2178 proto_tree_add_string(tree, id, tvb, start, length, value_ptr);
2181 proto_tree_add_string_format(tree, id, tvb, start, length, value_ptr,
2185 proto_tree_add_string_format_value(tree, id, tvb, start, length,
2186 value_ptr, format, ...);
2189 proto_tree_add_boolean(tree, id, tvb, start, length, value);
2192 proto_tree_add_boolean_format(tree, id, tvb, start, length, value,
2196 proto_tree_add_boolean_format_value(tree, id, tvb, start, length,
2197 value, format, ...);
2200 proto_tree_add_float(tree, id, tvb, start, length, value);
2203 proto_tree_add_float_format(tree, id, tvb, start, length, value,
2207 proto_tree_add_float_format_value(tree, id, tvb, start, length,
2208 value, format, ...);
2211 proto_tree_add_double(tree, id, tvb, start, length, value);
2214 proto_tree_add_double_format(tree, id, tvb, start, length, value,
2218 proto_tree_add_double_format_value(tree, id, tvb, start, length,
2219 value, format, ...);
2222 proto_tree_add_uint(tree, id, tvb, start, length, value);
2225 proto_tree_add_uint_format(tree, id, tvb, start, length, value,
2229 proto_tree_add_uint_format_value(tree, id, tvb, start, length,
2230 value, format, ...);
2233 proto_tree_add_uint64(tree, id, tvb, start, length, value);
2236 proto_tree_add_uint64_format(tree, id, tvb, start, length, value,
2240 proto_tree_add_uint64_format_value(tree, id, tvb, start, length,
2241 value, format, ...);
2244 proto_tree_add_int(tree, id, tvb, start, length, value);
2247 proto_tree_add_int_format(tree, id, tvb, start, length, value,
2251 proto_tree_add_int_format_value(tree, id, tvb, start, length,
2252 value, format, ...);
2255 proto_tree_add_int64(tree, id, tvb, start, length, value);
2258 proto_tree_add_int64_format(tree, id, tvb, start, length, value,
2262 proto_tree_add_int64_format_value(tree, id, tvb, start, length,
2263 value, format, ...);
2266 proto_tree_add_text(tree, tvb, start, length, format, ...);
2269 proto_tree_add_text_valist(tree, tvb, start, length, format, ap);
2272 proto_tree_add_guid(tree, id, tvb, start, length, value_ptr);
2275 proto_tree_add_guid_format(tree, id, tvb, start, length, value_ptr,
2279 proto_tree_add_guid_format_value(tree, id, tvb, start, length,
2280 value_ptr, format, ...);
2283 proto_tree_add_oid(tree, id, tvb, start, length, value_ptr);
2286 proto_tree_add_oid_format(tree, id, tvb, start, length, value_ptr,
2290 proto_tree_add_eui64(tree, id, tvb, start, length, value);
2293 proto_tree_add_eui64_format(tree, id, tvb, start, length, value,
2297 proto_tree_add_eui64_format_value(tree, id, tvb, start, length,
2298 value, format, ...);
2301 proto_tree_add_oid_format_value(tree, id, tvb, start, length,
2302 value_ptr, format, ...);
2305 proto_tree_add_bits_item(tree, id, tvb, bit_offset, no_of_bits,
2309 proto_tree_add_bits_ret_val(tree, id, tvb, bit_offset, no_of_bits,
2310 return_value, little_endian);
2313 proto_tree_add_bitmask(tree, tvb, start, header, ett, fields,
2317 proto_tree_add_bitmask_text(tree, tvb, offset, len, name, fallback,
2318 ett, fields, little_endian, flags);
2320 The 'tree' argument is the tree to which the item is to be added. The
2321 'tvb' argument is the tvbuff from which the item's value is being
2322 extracted; the 'start' argument is the offset from the beginning of that
2323 tvbuff of the item being added, and the 'length' argument is the length,
2324 in bytes, of the item, bit_offset is the offset in bits and no_of_bits
2325 is the length in bits.
2327 The length of some items cannot be determined until the item has been
2328 dissected; to add such an item, add it with a length of -1, and, when the
2329 dissection is complete, set the length with 'proto_item_set_len()':
2332 proto_item_set_len(ti, length);
2334 The "ti" argument is the value returned by the call that added the item
2335 to the tree, and the "length" argument is the length of the item.
2337 proto_tree_add_item()
2338 ---------------------
2339 proto_tree_add_item is used when you wish to do no special formatting.
2340 The item added to the GUI tree will contain the name (as passed in the
2341 proto_register_*() function) and a value. The value will be fetched
2342 from the tvbuff by proto_tree_add_item(), based on the type of the field
2343 and, for integral and Boolean fields, the byte order of the value; the
2344 byte order, for items for which that's relevant, is specified by the
2345 'encoding' argument, which is ENC_LITTLE_ENDIAN if the value is
2346 little-endian and ENC_BIG_ENDIAN if it is big-endian. If the byte order
2347 is not relevant, use ENC_NA (Not Applicable). For some field types,
2348 additional encoding information, such as the character encoding for
2349 character strings, are supported.
2351 Now that definitions of fields have detailed information about bitfield
2352 fields, you can use proto_tree_add_item() with no extra processing to
2353 add bitfield values to your tree. Here's an example. Take the Format
2354 Identifier (FID) field in the Transmission Header (TH) portion of the SNA
2355 protocol. The FID is the high nibble of the first byte of the TH. The
2356 FID would be registered like this:
2358 name = "Format Identifier"
2359 abbrev = "sna.th.fid"
2362 strings = sna_th_fid_vals
2365 The bitmask contains the value which would leave only the FID if bitwise-ANDed
2366 against the parent field, the first byte of the TH.
2368 The code to add the FID to the tree would be;
2370 proto_tree_add_item(bf_tree, hf_sna_th_fid, tvb, offset, 1,
2373 The definition of the field already has the information about bitmasking
2374 and bitshifting, so it does the work of masking and shifting for us!
2375 This also means that you no longer have to create value_string structs
2376 with the values bitshifted. The value_string for FID looks like this,
2377 even though the FID value is actually contained in the high nibble.
2378 (You'd expect the values to be 0x0, 0x10, 0x20, etc.)
2380 /* Format Identifier */
2381 static const value_string sna_th_fid_vals[] = {
2382 { 0x0, "SNA device <--> Non-SNA Device" },
2383 { 0x1, "Subarea Node <--> Subarea Node" },
2384 { 0x2, "Subarea Node <--> PU2" },
2385 { 0x3, "Subarea Node or SNA host <--> Subarea Node" },
2388 { 0xf, "Adjacent Subarea Nodes" },
2392 The final implication of this is that display filters work the way you'd
2393 naturally expect them to. You'd type "sna.th.fid == 0xf" to find Adjacent
2394 Subarea Nodes. The user does not have to shift the value of the FID to
2395 the high nibble of the byte ("sna.th.fid == 0xf0") as was necessary
2398 proto_tree_add_protocol_format()
2399 --------------------------------
2400 proto_tree_add_protocol_format is used to add the top-level item for the
2401 protocol when the dissector routine wants complete control over how the
2402 field and value will be represented on the GUI tree. The ID value for
2403 the protocol is passed in as the "id" argument; the rest of the
2404 arguments are a "printf"-style format and any arguments for that format.
2405 The caller must include the name of the protocol in the format; it is
2406 not added automatically as in proto_tree_add_item().
2408 proto_tree_add_none_format()
2409 ----------------------------
2410 proto_tree_add_none_format is used to add an item of type FT_NONE.
2411 The caller must include the name of the field in the format; it is
2412 not added automatically as in proto_tree_add_item().
2414 proto_tree_add_bytes()
2415 proto_tree_add_time()
2416 proto_tree_add_ipxnet()
2417 proto_tree_add_ipv4()
2418 proto_tree_add_ipv6()
2419 proto_tree_add_ether()
2420 proto_tree_add_string()
2421 proto_tree_add_boolean()
2422 proto_tree_add_float()
2423 proto_tree_add_double()
2424 proto_tree_add_uint()
2425 proto_tree_add_uint64()
2426 proto_tree_add_int()
2427 proto_tree_add_int64()
2428 proto_tree_add_guid()
2429 proto_tree_add_oid()
2430 proto_tree_add_eui64()
2431 ------------------------
2432 These routines are used to add items to the protocol tree if either:
2434 the value of the item to be added isn't just extracted from the
2435 packet data, but is computed from data in the packet;
2437 the value was fetched into a variable.
2439 The 'value' argument has the value to be added to the tree.
2441 NOTE: in all cases where the 'value' argument is a pointer, a copy is
2442 made of the object pointed to; if you have dynamically allocated a
2443 buffer for the object, that buffer will not be freed when the protocol
2444 tree is freed - you must free the buffer yourself when you don't need it
2447 For proto_tree_add_bytes(), the 'value_ptr' argument is a pointer to a
2450 For proto_tree_add_bytes_format() and proto_tree_add_bytes_format_value(), the
2451 'value_ptr' argument is a pointer to a sequence of bytes or NULL if the bytes
2452 should be taken from the given TVB using the given offset and length.
2454 For proto_tree_add_time(), the 'value_ptr' argument is a pointer to an
2455 "nstime_t", which is a structure containing the time to be added; it has
2456 'secs' and 'nsecs' members, giving the integral part and the fractional
2457 part of a time in units of seconds, with 'nsecs' being the number of
2458 nanoseconds. For absolute times, "secs" is a UNIX-style seconds since
2459 January 1, 1970, 00:00:00 GMT value.
2461 For proto_tree_add_ipxnet(), the 'value' argument is a 32-bit IPX
2464 For proto_tree_add_ipv4(), the 'value' argument is a 32-bit IPv4
2465 address, in network byte order.
2467 For proto_tree_add_ipv6(), the 'value_ptr' argument is a pointer to a
2468 128-bit IPv6 address.
2470 For proto_tree_add_ether(), the 'value_ptr' argument is a pointer to a
2473 For proto_tree_add_string(), the 'value_ptr' argument is a pointer to a
2476 For proto_tree_add_boolean(), the 'value' argument is a 32-bit integer.
2477 It is masked and shifted as defined by the field info after which zero
2478 means "false", and non-zero means "true".
2480 For proto_tree_add_float(), the 'value' argument is a 'float' in the
2481 host's floating-point format.
2483 For proto_tree_add_double(), the 'value' argument is a 'double' in the
2484 host's floating-point format.
2486 For proto_tree_add_uint(), the 'value' argument is a 32-bit unsigned
2487 integer value, in host byte order. (This routine cannot be used to add
2490 For proto_tree_add_uint64(), the 'value' argument is a 64-bit unsigned
2491 integer value, in host byte order.
2493 For proto_tree_add_int(), the 'value' argument is a 32-bit signed
2494 integer value, in host byte order. (This routine cannot be used to add
2497 For proto_tree_add_int64(), the 'value' argument is a 64-bit signed
2498 integer value, in host byte order.
2500 For proto_tree_add_guid(), the 'value_ptr' argument is a pointer to an
2503 For proto_tree_add_oid(), the 'value_ptr' argument is a pointer to an
2504 ASN.1 Object Identifier.
2506 For proto_tree_add_eui64(), the 'value' argument is a 64-bit integer
2509 proto_tree_add_bytes_format()
2510 proto_tree_add_time_format()
2511 proto_tree_add_ipxnet_format()
2512 proto_tree_add_ipv4_format()
2513 proto_tree_add_ipv6_format()
2514 proto_tree_add_ether_format()
2515 proto_tree_add_string_format()
2516 proto_tree_add_boolean_format()
2517 proto_tree_add_float_format()
2518 proto_tree_add_double_format()
2519 proto_tree_add_uint_format()
2520 proto_tree_add_uint64_format()
2521 proto_tree_add_int_format()
2522 proto_tree_add_int64_format()
2523 proto_tree_add_guid_format()
2524 proto_tree_add_oid_format()
2525 proto_tree_add_eui64_format()
2526 ----------------------------
2527 These routines are used to add items to the protocol tree when the
2528 dissector routine wants complete control over how the field and value
2529 will be represented on the GUI tree. The argument giving the value is
2530 the same as the corresponding proto_tree_add_XXX() function; the rest of
2531 the arguments are a "printf"-style format and any arguments for that
2532 format. The caller must include the name of the field in the format; it
2533 is not added automatically as in the proto_tree_add_XXX() functions.
2535 proto_tree_add_bytes_format_value()
2536 proto_tree_add_time_format_value()
2537 proto_tree_add_ipxnet_format_value()
2538 proto_tree_add_ipv4_format_value()
2539 proto_tree_add_ipv6_format_value()
2540 proto_tree_add_ether_format_value()
2541 proto_tree_add_string_format_value()
2542 proto_tree_add_boolean_format_value()
2543 proto_tree_add_float_format_value()
2544 proto_tree_add_double_format_value()
2545 proto_tree_add_uint_format_value()
2546 proto_tree_add_uint64_format_value()
2547 proto_tree_add_int_format_value()
2548 proto_tree_add_int64_format_value()
2549 proto_tree_add_guid_format_value()
2550 proto_tree_add_oid_format_value()
2551 proto_tree_add_eui64_format_value()
2552 ------------------------------------
2554 These routines are used to add items to the protocol tree when the
2555 dissector routine wants complete control over how the value will be
2556 represented on the GUI tree. The argument giving the value is the same
2557 as the corresponding proto_tree_add_XXX() function; the rest of the
2558 arguments are a "printf"-style format and any arguments for that format.
2559 With these routines, unlike the proto_tree_add_XXX_format() routines,
2560 the name of the field is added automatically as in the
2561 proto_tree_add_XXX() functions; only the value is added with the format.
2563 proto_tree_add_text()
2564 ---------------------
2565 proto_tree_add_text() is used to add a label to the GUI tree. It will
2566 contain no value, so it is not searchable in the display filter process.
2567 This function was needed in the transition from the old-style proto_tree
2568 to this new-style proto_tree so that Wireshark would still decode all
2569 protocols w/o being able to filter on all protocols and fields.
2570 Otherwise we would have had to cripple Wireshark's functionality while we
2571 converted all the old-style proto_tree calls to the new-style proto_tree
2572 calls. In other words, you should not use this in new code unless you've got
2573 a specific reason (see below).
2575 This can also be used for items with subtrees, which may not have values
2576 themselves - the items in the subtree are the ones with values.
2578 For a subtree, the label on the subtree might reflect some of the items
2579 in the subtree. This means the label can't be set until at least some
2580 of the items in the subtree have been dissected. To do this, use
2581 'proto_item_set_text()' or 'proto_item_append_text()':
2584 proto_item_set_text(proto_item *ti, ...);
2587 proto_item_append_text(proto_item *ti, ...);
2589 'proto_item_set_text()' takes as an argument the value returned by
2590 'proto_tree_add_text()', a 'printf'-style format string, and a set of
2591 arguments corresponding to '%' format items in that string, and replaces
2592 the text for the item created by 'proto_tree_add_text()' with the result
2593 of applying the arguments to the format string.
2595 'proto_item_append_text()' is similar, but it appends to the text for
2596 the item the result of applying the arguments to the format string.
2598 For example, early in the dissection, one might do:
2600 ti = proto_tree_add_text(tree, tvb, offset, length, <label>);
2604 proto_item_set_text(ti, "%s: %s", type, value);
2606 after the "type" and "value" fields have been extracted and dissected.
2607 <label> would be a label giving what information about the subtree is
2608 available without dissecting any of the data in the subtree.
2610 Note that an exception might be thrown when trying to extract the values of
2611 the items used to set the label, if not all the bytes of the item are
2612 available. Thus, one should create the item with text that is as
2613 meaningful as possible, and set it or append additional information to
2614 it as the values needed to supply that information are extracted.
2616 proto_tree_add_text_valist()
2617 ----------------------------
2618 This is like proto_tree_add_text(), but takes, as the last argument, a
2619 'va_list'; it is used to allow routines that take a printf-like
2620 variable-length list of arguments to add a text item to the protocol
2623 proto_tree_add_bits_item()
2624 --------------------------
2625 Adds a number of bits to the protocol tree which does not have to be byte
2626 aligned. The offset and length is in bits.
2629 ..10 1010 10.. .... "value" (formatted as FT_ indicates).
2631 proto_tree_add_bits_ret_val()
2632 -----------------------------
2633 Works in the same way but also returns the value of the read bits.
2635 proto_tree_add_bitmask() and proto_tree_add_bitmask_text()
2636 ----------------------------------------------------------
2637 This function provides an easy to use and convenient helper function
2638 to manage many types of common bitmasks that occur in protocols.
2640 This function will dissect a 1/2/3/4 byte large bitmask into its individual
2642 header is an integer type and must be of type FT_[U]INT{8|16|24|32} and
2643 represents the entire width of the bitmask.
2645 'header' and 'ett' are the hf fields and ett field respectively to create an
2646 expansion that covers the 1-4 bytes of the bitmask.
2648 'fields' is a NULL terminated array of pointers to hf fields representing
2649 the individual subfields of the bitmask. These fields must either be integers
2650 of the same byte width as 'header' or of the type FT_BOOLEAN.
2651 Each of the entries in 'fields' will be dissected as an item under the
2652 'header' expansion and also IF the field is a boolean and IF it is set to 1,
2653 then the name of that boolean field will be printed on the 'header' expansion
2654 line. For integer type subfields that have a value_string defined, the
2655 matched string from that value_string will be printed on the expansion line
2658 Example: (from the SCSI dissector)
2659 static int hf_scsi_inq_peripheral = -1;
2660 static int hf_scsi_inq_qualifier = -1;
2661 static int hf_scsi_inq_devtype = -1;
2663 static gint ett_scsi_inq_peripheral = -1;
2665 static const int *peripheal_fields[] = {
2666 &hf_scsi_inq_qualifier,
2667 &hf_scsi_inq_devtype,
2671 /* Qualifier and DeviceType */
2672 proto_tree_add_bitmask(tree, tvb, offset, hf_scsi_inq_peripheral,
2673 ett_scsi_inq_peripheral, peripheal_fields, FALSE);
2676 { &hf_scsi_inq_peripheral,
2677 {"Peripheral", "scsi.inquiry.preipheral", FT_UINT8, BASE_HEX,
2678 NULL, 0, NULL, HFILL}},
2679 { &hf_scsi_inq_qualifier,
2680 {"Qualifier", "scsi.inquiry.qualifier", FT_UINT8, BASE_HEX,
2681 VALS (scsi_qualifier_val), 0xE0, NULL, HFILL}},
2682 { &hf_scsi_inq_devtype,
2683 {"Device Type", "scsi.inquiry.devtype", FT_UINT8, BASE_HEX,
2684 VALS (scsi_devtype_val), SCSI_DEV_BITS, NULL, HFILL}},
2687 Which provides very pretty dissection of this one byte bitmask.
2689 Peripheral: 0x05, Qualifier: Device type is connected to logical unit, Device Type: CD-ROM
2690 000. .... = Qualifier: Device type is connected to logical unit (0x00)
2691 ...0 0101 = Device Type: CD-ROM (0x05)
2693 The proto_tree_add_bitmask_text() function is an extended version of
2694 the proto_tree_add_bitmask() function. In addition, it allows to:
2695 - Provide a leading text (e.g. "Flags: ") that will appear before
2696 the comma-separated list of field values
2697 - Provide a fallback text (e.g. "None") that will be appended if
2698 no fields warranted a change to the top-level title.
2699 - Using flags, specify which fields will affect the top-level title.
2701 There are the following flags defined:
2703 BMT_NO_APPEND - the title is taken "as-is" from the 'name' argument.
2704 BMT_NO_INT - only boolean flags are added to the title.
2705 BMT_NO_FALSE - boolean flags are only added to the title if they are set.
2706 BMT_NO_TFS - only add flag name to the title, do not use true_false_string
2708 The proto_tree_add_bitmask() behavior can be obtained by providing
2709 both 'name' and 'fallback' arguments as NULL, and a flags of
2710 (BMT_NO_FALSE|BMT_NO_TFS).
2712 PROTO_ITEM_SET_GENERATED()
2713 --------------------------
2714 PROTO_ITEM_SET_GENERATED is used to mark fields as not being read from the
2715 captured data directly, but inferred from one or more values.
2717 One of the primary uses of this is the presentation of verification of
2718 checksums. Every IP packet has a checksum line, which can present the result
2719 of the checksum verification, if enabled in the preferences. The result is
2720 presented as a subtree, where the result is enclosed in square brackets
2721 indicating a generated field.
2723 Header checksum: 0x3d42 [correct]
2727 PROTO_ITEM_SET_HIDDEN()
2728 -----------------------
2729 PROTO_ITEM_SET_HIDDEN is used to hide fields, which have already been added
2730 to the tree, from being visible in the displayed tree.
2732 NOTE that creating hidden fields is actually quite a bad idea from a UI design
2733 perspective because the user (someone who did not write nor has ever seen the
2734 code) has no way of knowing that hidden fields are there to be filtered on
2735 thus defeating the whole purpose of putting them there. A Better Way might
2736 be to add the fields (that might otherwise be hidden) to a subtree where they
2737 won't be seen unless the user opens the subtree--but they can be found if the
2740 One use for hidden fields (which would be better implemented using visible
2741 fields in a subtree) follows: The caller may want a value to be
2742 included in a tree so that the packet can be filtered on this field, but
2743 the representation of that field in the tree is not appropriate. An
2744 example is the token-ring routing information field (RIF). The best way
2745 to show the RIF in a GUI is by a sequence of ring and bridge numbers.
2746 Rings are 3-digit hex numbers, and bridges are single hex digits:
2748 RIF: 001-A-013-9-C0F-B-555
2750 In the case of RIF, the programmer should use a field with no value and
2751 use proto_tree_add_none_format() to build the above representation. The
2752 programmer can then add the ring and bridge values, one-by-one, with
2753 proto_tree_add_item() and hide them with PROTO_ITEM_SET_HIDDEN() so that the
2754 user can then filter on or search for a particular ring or bridge. Here's a
2755 skeleton of how the programmer might code this.
2758 rif = create_rif_string(...);
2760 proto_tree_add_none_format(tree, hf_tr_rif_label, ..., "RIF: %s", rif);
2762 for(i = 0; i < num_rings; i++) {
2765 pi = proto_tree_add_item(tree, hf_tr_rif_ring, ...,
2767 PROTO_ITEM_SET_HIDDEN(pi);
2769 for(i = 0; i < num_rings - 1; i++) {
2772 pi = proto_tree_add_item(tree, hf_tr_rif_bridge, ...,
2774 PROTO_ITEM_SET_HIDDEN(pi);
2777 The logical tree has these items:
2779 hf_tr_rif_label, text="RIF: 001-A-013-9-C0F-B-555", value = NONE
2780 hf_tr_rif_ring, hidden, value=0x001
2781 hf_tr_rif_bridge, hidden, value=0xA
2782 hf_tr_rif_ring, hidden, value=0x013
2783 hf_tr_rif_bridge, hidden, value=0x9
2784 hf_tr_rif_ring, hidden, value=0xC0F
2785 hf_tr_rif_bridge, hidden, value=0xB
2786 hf_tr_rif_ring, hidden, value=0x555
2788 GUI or print code will not display the hidden fields, but a display
2789 filter or "packet grep" routine will still see the values. The possible
2790 filter is then possible:
2792 tr.rif_ring eq 0x013
2796 PROTO_ITEM_SET_URL is used to mark fields as containing a URL. This can only
2797 be done with fields of type FT_STRING(Z). If these fields are presented they
2798 are underlined, as could be done in a browser. These fields are sensitive to
2799 clicks as well, launching the configured browser with this URL as parameter.
2801 1.7 Utility routines.
2803 1.7.1 match_strval, match_strval_ext, val_to_str and val_to_str_ext.
2805 A dissector may need to convert a value to a string, using a
2806 'value_string' structure, by hand, rather than by declaring a field with
2807 an associated 'value_string' structure; this might be used, for example,
2808 to generate a COL_INFO line for a frame.
2810 'match_strval()' will do that:
2813 match_strval(guint32 val, const value_string *vs)
2815 It will look up the value 'val' in the 'value_string' table pointed to
2816 by 'vs', and return either the corresponding string, or NULL if the
2817 value could not be found in the table. Note that, unless 'val' is
2818 guaranteed to be a value in the 'value_string' table ("guaranteed" as in
2819 "the code has already checked that it's one of those values" or "the
2820 table handles all possible values of the size of 'val'", not "the
2821 protocol spec says it has to be" - protocol specs do not prevent invalid
2822 packets from being put onto a network or into a purported packet capture
2823 file), you must check whether 'match_strval()' returns NULL, and arrange
2824 that its return value not be dereferenced if it's NULL. 'val_to_str()'
2825 can be used to generate a string for values not found in the table:
2828 val_to_str(guint32 val, const value_string *vs, const char *fmt)
2830 If the value 'val' is found in the 'value_string' table pointed to by
2831 'vs', 'val_to_str' will return the corresponding string; otherwise, it
2832 will use 'fmt' as an 'sprintf'-style format, with 'val' as an argument,
2833 to generate a string, and will return a pointer to that string.
2834 You can use it in a call to generate a COL_INFO line for a frame such as
2836 col_add_fstr(COL_INFO, ", %s", val_to_str(val, table, "Unknown %d"));
2838 The match_strval_ext and val_to_str_ext functions are "extended" versions
2839 of match_strval and val_to_str. They should be used for large value-string
2840 arrays which contain many entries. They implement value to string conversions
2841 which will do either a direct access or a binary search of the
2842 value string array if possible. See "Extended Value Strings" under
2843 section 1.6 "Constructing the protocol tree" for more information.
2845 See epan/value_string.h for detailed information on the various value_string
2849 1.7.2 match_strrval and rval_to_str.
2851 A dissector may need to convert a range of values to a string, using a
2852 'range_string' structure.
2854 'match_strrval()' will do that:
2857 match_strrval(guint32 val, const range_string *rs)
2859 It will look up the value 'val' in the 'range_string' table pointed to
2860 by 'rs', and return either the corresponding string, or NULL if the
2861 value could not be found in the table. Please note that its base
2862 behavior is inherited from match_strval().
2864 'rval_to_str()' can be used to generate a string for values not found in
2868 rval_to_str(guint32 val, const range_string *rs, const char *fmt)
2870 If the value 'val' is found in the 'range_string' table pointed to by
2871 'rs', 'rval_to_str' will return the corresponding string; otherwise, it
2872 will use 'fmt' as an 'sprintf'-style format, with 'val' as an argument,
2873 to generate a string, and will return a pointer to that string. Please
2874 note that its base behavior is inherited from match_strval().
2876 1.8 Calling Other Dissectors.
2878 As each dissector completes its portion of the protocol analysis, it
2879 is expected to create a new tvbuff of type TVBUFF_SUBSET which
2880 contains the payload portion of the protocol (that is, the bytes
2881 that are relevant to the next dissector).
2883 The syntax for creating a new TVBUFF_SUBSET is:
2885 next_tvb = tvb_new_subset(tvb, offset, length, reported_length)
2888 tvb is the tvbuff that the dissector has been working on. It
2889 can be a tvbuff of any type.
2891 next_tvb is the new TVBUFF_SUBSET.
2893 offset is the byte offset of 'tvb' at which the new tvbuff
2894 should start. The first byte is the 0th byte.
2896 length is the number of bytes in the new TVBUFF_SUBSET. A length
2897 argument of -1 says to use as many bytes as are available in
2900 reported_length is the number of bytes that the current protocol
2901 says should be in the payload. A reported_length of -1 says that
2902 the protocol doesn't say anything about the size of its payload.
2905 An example from packet-ipx.c -
2908 dissect_ipx(tvbuff_t *tvb, packet_info *pinfo, proto_tree *tree)
2911 int reported_length, available_length;
2914 /* Make the next tvbuff */
2916 /* IPX does have a length value in the header, so calculate report_length */
2917 Set this to -1 if there isn't any length information in the protocol
2919 reported_length = ipx_length - IPX_HEADER_LEN;
2921 /* Calculate the available data in the packet,
2922 set this to -1 to use all the data in the tv_buffer
2924 available_length = tvb_length(tvb) - IPX_HEADER_LEN;
2926 /* Create the tvbuffer for the next dissector */
2927 next_tvb = tvb_new_subset(tvb, IPX_HEADER_LEN,
2928 MIN(available_length, reported_length),
2931 /* call the next dissector */
2932 dissector_next( next_tvb, pinfo, tree);
2935 1.9 Editing Makefile.common and CMakeLists.txt to add your dissector.
2937 To arrange that your dissector will be built as part of Wireshark, you
2938 must add the name of the source file for your dissector to the
2939 'DISSECTOR_SRC' macro in the 'Makefile.common' file in the 'epan/dissectors'
2940 directory. (Note that this is for modern versions of UNIX, so there
2941 is no 14-character limitation on file names, and for modern versions of
2942 Windows, so there is no 8.3-character limitation on file names.)
2944 If your dissector also has its own header file or files, you must add
2945 them to the 'DISSECTOR_INCLUDES' macro in the 'Makefile.common' file in
2946 the 'epan/dissectors' directory, so that it's included when release source
2947 tarballs are built (otherwise, the source in the release tarballs won't
2950 In addition to the above, you should add your dissector source file name
2951 to the DISSECTOR_SRC section of epan/CMakeLists.txt
2954 1.10 Using the SVN source code tree.
2956 See <http://www.wireshark.org/develop.html>
2959 1.10a Using git with the SVN source code tree.
2961 Install git and the git-svn package.
2962 Run "mkdir git; cd git; git svn clone <svn-url>", e.g. if you are using
2963 the anonymous svn tree, run
2964 "git svn clone http://anonsvn.wireshark.org/wireshark/trunk/"
2966 After that, a typical workflow may look like this (from "man git-svn"):
2968 # Clone a repo (like git clone):
2969 git svn clone http://svn.example.com/project/trunk
2970 # Enter the newly cloned directory:
2972 # You should be on master branch, double-check with ´git branch´
2974 # Do some work and commit locally to git:
2976 # Something is committed to SVN, rebase your local changes against the
2977 # latest changes in SVN:
2979 # Now commit your changes (that were committed previously using git) to SVN
2980 # as well as automatically updating your working HEAD:
2982 # Append svn:ignore settings to the default git exclude file:
2983 git svn show-ignore >> .git/info/exclude
2986 1.11 Submitting code for your new dissector.
2988 - VERIFY that your dissector code does not use prohibited or deprecated APIs
2990 perl <wireshark_root>/tools/checkAPIs.pl <source-filename(s)>
2992 - TEST YOUR DISSECTOR BEFORE SUBMITTING IT.
2993 Use fuzz-test.sh and/or randpkt against your dissector. These are
2994 described at <http://wiki.wireshark.org/FuzzTesting>.
2996 - Subscribe to <mailto:wireshark-dev[AT]wireshark.org> by sending an email to
2997 <mailto:wireshark-dev-request[AT]wireshark.org?body="help"> or visiting
2998 <http://www.wireshark.org/lists/>.
3000 - 'svn add' all the files of your new dissector.
3002 - 'svn diff' the workspace and save the result to a file.
3004 - Edit the diff file - remove any changes unrelated to your new dissector,
3005 e.g. changes in config.nmake
3007 - Submit a bug report to the Wireshark bug database, found at
3008 <http://bugs.wireshark.org>, qualified as an enhancement and attach your
3009 diff file there. Set the review request flag to '?' so it will pop up in
3010 the patch review list.
3012 - Create a Wiki page on the protocol at <http://wiki.wireshark.org>.
3013 A template is provided so it is easy to setup in a consistent style.
3014 See: <http://wiki.wireshark.org/HowToEdit>
3015 and <http://wiki.wireshark.org/ProtocolReference>
3017 - If possible, add sample capture files to the sample captures page at
3018 <http://wiki.wireshark.org/SampleCaptures>. These files are used by
3019 the automated build system for fuzz testing.
3021 - If you find that you are contributing a lot to wireshark on an ongoing
3022 basis you can request to become a committer which will allow you to
3023 commit files to subversion directly.
3025 2. Advanced dissector topics.
3029 Some of the advanced features are being worked on constantly. When using them
3030 it is wise to check the relevant header and source files for additional details.
3032 2.2 Following "conversations".
3034 In wireshark a conversation is defined as a series of data packets between two
3035 address:port combinations. A conversation is not sensitive to the direction of
3036 the packet. The same conversation will be returned for a packet bound from
3037 ServerA:1000 to ClientA:2000 and the packet from ClientA:2000 to ServerA:1000.
3039 2.2.1 Conversation Routines
3041 There are six routines that you will use to work with a conversation:
3042 conversation_new, find_conversation, conversation_add_proto_data,
3043 conversation_get_proto_data, conversation_delete_proto_data,
3044 and conversation_set_dissector.
3047 2.2.1.1 The conversation_init function.
3049 This is an internal routine for the conversation code. As such you
3050 will not have to call this routine. Just be aware that this routine is
3051 called at the start of each capture and before the packets are filtered
3052 with a display filter. The routine will destroy all stored
3053 conversations. This routine does NOT clean up any data pointers that are
3054 passed in the conversation_add_proto_data 'data' variable. You are
3055 responsible for this clean up if you pass a malloc'ed pointer
3058 See item 2.2.1.5 for more information about use of the 'data' pointer.
3061 2.2.1.2 The conversation_new function.
3063 This routine will create a new conversation based upon two address/port
3064 pairs. If you want to associate with the conversation a pointer to a
3065 private data structure you must use the conversation_add_proto_data
3066 function. The ptype variable is used to differentiate between
3067 conversations over different protocols, i.e. TCP and UDP. The options
3068 variable is used to define a conversation that will accept any destination
3069 address and/or port. Set options = 0 if the destination port and address
3070 are know when conversation_new is called. See section 2.4 for more
3071 information on usage of the options parameter.
3073 The conversation_new prototype:
3074 conversation_t *conversation_new(guint32 setup_frame, address *addr1,
3075 address *addr2, port_type ptype, guint32 port1, guint32 port2,
3079 guint32 setup_frame = The lowest numbered frame for this conversation
3080 address* addr1 = first data packet address
3081 address* addr2 = second data packet address
3082 port_type ptype = port type, this is defined in packet.h
3083 guint32 port1 = first data packet port
3084 guint32 port2 = second data packet port
3085 guint options = conversation options, NO_ADDR2 and/or NO_PORT2
3087 setup_frame indicates the first frame for this conversation, and is used to
3088 distinguish multiple conversations with the same addr1/port1 and addr2/port2
3089 pair that occur within the same capture session.
3091 "addr1" and "port1" are the first address/port pair; "addr2" and "port2"
3092 are the second address/port pair. A conversation doesn't have source
3093 and destination address/port pairs - packets in a conversation go in
3094 both directions - so "addr1"/"port1" may be the source or destination
3095 address/port pair; "addr2"/"port2" would be the other pair.
3097 If NO_ADDR2 is specified, the conversation is set up so that a
3098 conversation lookup will match only the "addr1" address; if NO_PORT2 is
3099 specified, the conversation is set up so that a conversation lookup will
3100 match only the "port1" port; if both are specified, i.e.
3101 NO_ADDR2|NO_PORT2, the conversation is set up so that the lookup will
3102 match only the "addr1"/"port1" address/port pair. This can be used if a
3103 packet indicates that, later in the capture, a conversation will be
3104 created using certain addresses and ports, in the case where the packet
3105 doesn't specify the addresses and ports of both sides.
3107 2.2.1.3 The find_conversation function.
3109 Call this routine to look up a conversation. If no conversation is found,
3110 the routine will return a NULL value.
3112 The find_conversation prototype:
3114 conversation_t *find_conversation(guint32 frame_num, address *addr_a,
3115 address *addr_b, port_type ptype, guint32 port_a, guint32 port_b,
3119 guint32 frame_num = a frame number to match
3120 address* addr_a = first address
3121 address* addr_b = second address
3122 port_type ptype = port type
3123 guint32 port_a = first data packet port
3124 guint32 port_b = second data packet port
3125 guint options = conversation options, NO_ADDR_B and/or NO_PORT_B
3127 frame_num is a frame number to match. The conversation returned is where
3128 (frame_num >= conversation->setup_frame
3129 && frame_num < conversation->next->setup_frame)
3130 Suppose there are a total of 3 conversations (A, B, and C) that match
3131 addr_a/port_a and addr_b/port_b, where the setup_frame used in
3132 conversation_new() for A, B and C are 10, 50, and 100 respectively. The
3133 frame_num passed in find_conversation is compared to the setup_frame of each
3134 conversation. So if (frame_num >= 10 && frame_num < 50), conversation A is
3135 returned. If (frame_num >= 50 && frame_num < 100), conversation B is returned.
3136 If (frame_num >= 100) conversation C is returned.
3138 "addr_a" and "port_a" are the first address/port pair; "addr_b" and
3139 "port_b" are the second address/port pair. Again, as a conversation
3140 doesn't have source and destination address/port pairs, so
3141 "addr_a"/"port_a" may be the source or destination address/port pair;
3142 "addr_b"/"port_b" would be the other pair. The search will match the
3143 "a" address/port pair against both the "1" and "2" address/port pairs,
3144 and match the "b" address/port pair against both the "2" and "1"
3145 address/port pairs; you don't have to worry about which side the "a" or
3146 "b" pairs correspond to.
3148 If the NO_ADDR_B flag was specified to "find_conversation()", the
3149 "addr_b" address will be treated as matching any "wildcarded" address;
3150 if the NO_PORT_B flag was specified, the "port_b" port will be treated
3151 as matching any "wildcarded" port. If both flags are specified, i.e.
3152 NO_ADDR_B|NO_PORT_B, the "addr_b" address will be treated as matching
3153 any "wildcarded" address and the "port_b" port will be treated as
3154 matching any "wildcarded" port.
3157 2.2.1.4 The find_or_create_conversation function.
3159 This convenience function will create find an existing conversation (by calling
3160 find_conversation()) and, if a conversation does not already exist, create a
3161 new conversation by calling conversation_new().
3163 The find_or_create_conversation prototype:
3165 extern conversation_t *find_or_create_conversation(packet_info *pinfo);
3168 packet_info *pinfo = the packet_info structure
3170 The frame number and the addresses necessary for find_conversation() and
3171 conversation_new() are taken from the pinfo structure (as is commonly done)
3172 and no 'options' are used.
3175 2.2.1.5 The conversation_add_proto_data function.
3177 Once you have created a conversation with conversation_new, you can
3178 associate data with it using this function.
3180 The conversation_add_proto_data prototype:
3182 void conversation_add_proto_data(conversation_t *conv, int proto,
3186 conversation_t *conv = the conversation in question
3187 int proto = registered protocol number
3188 void *data = dissector data structure
3190 "conversation" is the value returned by conversation_new. "proto" is a
3191 unique protocol number created with proto_register_protocol. Protocols
3192 are typically registered in the proto_register_XXXX section of your
3193 dissector. "data" is a pointer to the data you wish to associate with the
3194 conversation. "data" usually points to "se_alloc'd" memory; the
3195 memory will be automatically freed each time a new dissection begins
3196 and thus need not be managed (freed) by the dissector.
3197 Using the protocol number allows several dissectors to
3198 associate data with a given conversation.
3201 2.2.1.6 The conversation_get_proto_data function.
3203 After you have located a conversation with find_conversation, you can use
3204 this function to retrieve any data associated with it.
3206 The conversation_get_proto_data prototype:
3208 void *conversation_get_proto_data(conversation_t *conv, int proto);
3211 conversation_t *conv = the conversation in question
3212 int proto = registered protocol number
3214 "conversation" is the conversation created with conversation_new. "proto"
3215 is a unique protocol number created with proto_register_protocol,
3216 typically in the proto_register_XXXX portion of a dissector. The function
3217 returns a pointer to the data requested, or NULL if no data was found.
3220 2.2.1.7 The conversation_delete_proto_data function.
3222 After you are finished with a conversation, you can remove your association
3223 with this function. Please note that ONLY the conversation entry is
3224 removed. If you have allocated any memory for your data (other than with se_alloc),
3225 you must free it as well.
3227 The conversation_delete_proto_data prototype:
3229 void conversation_delete_proto_data(conversation_t *conv, int proto);
3232 conversation_t *conv = the conversation in question
3233 int proto = registered protocol number
3235 "conversation" is the conversation created with conversation_new. "proto"
3236 is a unique protocol number created with proto_register_protocol,
3237 typically in the proto_register_XXXX portion of a dissector.
3239 2.2.1.8 The conversation_set_dissector function
3241 This function sets the protocol dissector to be invoked whenever
3242 conversation parameters (addresses, port_types, ports, etc) are matched
3243 during the dissection of a packet.
3245 The conversation_set_dissector prototype:
3247 void conversation_set_dissector(conversation_t *conversation, const dissector_handle_t handle);
3250 conversation_t *conv = the conversation in question
3251 const dissector_handle_t handle = the dissector handle.
3254 2.2.2 Using timestamps relative to the conversation
3256 There is a framework to calculate timestamps relative to the start of the
3257 conversation. First of all the timestamp of the first packet that has been
3258 seen in the conversation must be kept in the protocol data to be able
3259 to calculate the timestamp of the current packet relative to the start
3260 of the conversation. The timestamp of the last packet that was seen in the
3261 conversation should also be kept in the protocol data. This way the
3262 delta time between the current packet and the previous packet in the
3263 conversation can be calculated.
3265 So add the following items to the struct that is used for the protocol data:
3270 The ts_prev value should only be set during the first run through the
3271 packets (ie pinfo->fd->flags.visited is false).
3273 Next step is to use the per-packet information (described in section 2.5)
3274 to keep the calculated delta timestamp, as it can only be calculated
3275 on the first run through the packets. This is because a packet can be
3276 selected in random order once the whole file has been read.
3278 After calculating the conversation timestamps, it is time to put them in
3279 the appropriate columns with the function 'col_set_time' (described in
3280 section 1.5.9). There are two columns for conversation timestamps:
3282 COL_REL_CONV_TIME, /* Relative time to beginning of conversation */
3283 COL_DELTA_CONV_TIME,/* Delta time to last frame in conversation */
3285 Last but not least, there MUST be a preference in each dissector that
3286 uses conversation timestamps that makes it possible to enable and
3287 disable the calculation of conversation timestamps. The main argument
3288 for this is that a higher level conversation is able to overwrite
3289 the values of lowel level conversations in these two columns. Being
3290 able to actively select which protocols may overwrite the conversation
3291 timestamp columns gives the user the power to control these columns.
3292 (A second reason is that conversation timestamps use the per-packet
3293 data structure which uses additional memory, which should be avoided
3294 if these timestamps are not needed)
3296 Have a look at the differences to packet-tcp.[ch] in SVN 22966 and
3297 SVN 23058 to see the implementation of conversation timestamps for
3301 2.2.3 The example conversation code using se_alloc'd memory.
3303 For a conversation between two IP addresses and ports you can use this as an
3304 example. This example uses se_alloc() to allocate memory and stores the data
3305 pointer in the conversation 'data' variable.
3307 /************************ Global values ************************/
3309 /* define your structure here */
3314 /* Registered protocol number */
3315 static int my_proto = -1;
3317 /********************* in the dissector routine *********************/
3319 /* the local variables in the dissector */
3321 conversation_t *conversation;
3322 my_entry_t *data_ptr;
3325 /* look up the conversation */
3327 conversation = find_conversation(pinfo->fd->num, &pinfo->src, &pinfo->dst,
3328 pinfo->ptype, pinfo->srcport, pinfo->destport, 0);
3330 /* if conversation found get the data pointer that you stored */
3332 data_ptr = (my_entry_t*)conversation_get_proto_data(conversation, my_proto);
3335 /* new conversation create local data structure */
3337 data_ptr = se_alloc(sizeof(my_entry_t));
3339 /*** add your code here to setup the new data structure ***/
3341 /* create the conversation with your data pointer */
3343 conversation = conversation_new(pinfo->fd->num, &pinfo->src, &pinfo->dst, pinfo->ptype,
3344 pinfo->srcport, pinfo->destport, 0);
3345 conversation_add_proto_data(conversation, my_proto, (void *)data_ptr);
3348 /* at this point the conversation data is ready */
3350 /***************** in the protocol register routine *****************/
3352 my_proto = proto_register_protocol("My Protocol", "My Protocol", "my_proto");
3355 2.2.4 An example conversation code that starts at a specific frame number.
3357 Sometimes a dissector has determined that a new conversation is needed that
3358 starts at a specific frame number, when a capture session encompasses multiple
3359 conversation that reuse the same src/dest ip/port pairs. You can use the
3360 conversation->setup_frame returned by find_conversation with
3361 pinfo->fd->num to determine whether or not there already exists a conversation
3362 that starts at the specific frame number.
3364 /* in the dissector routine */
3366 conversation = find_conversation(pinfo->fd->num, &pinfo->src, &pinfo->dst,
3367 pinfo->ptype, pinfo->srcport, pinfo->destport, 0);
3368 if (conversation == NULL || (conversation->setup_frame != pinfo->fd->num)) {
3369 /* It's not part of any conversation or the returned
3370 * conversation->setup_frame doesn't match the current frame
3373 conversation = conversation_new(pinfo->fd->num, &pinfo->src,
3374 &pinfo->dst, pinfo->ptype, pinfo->srcport, pinfo->destport,
3379 2.2.5 The example conversation code using conversation index field.
3381 Sometimes the conversation isn't enough to define a unique data storage
3382 value for the network traffic. For example if you are storing information
3383 about requests carried in a conversation, the request may have an
3384 identifier that is used to define the request. In this case the
3385 conversation and the identifier are required to find the data storage
3386 pointer. You can use the conversation data structure index value to
3387 uniquely define the conversation.
3389 See packet-afs.c for an example of how to use the conversation index. In
3390 this dissector multiple requests are sent in the same conversation. To store
3391 information for each request the dissector has an internal hash table based
3392 upon the conversation index and values inside the request packets.
3395 /* in the dissector routine */
3397 /* to find a request value, first lookup conversation to get index */
3398 /* then used the conversation index, and request data to find data */
3399 /* in the local hash table */
3401 conversation = find_or_create_conversation(pinfo);
3403 request_key.conversation = conversation->index;
3404 request_key.service = pntohs(&rxh->serviceId);
3405 request_key.callnumber = pntohl(&rxh->callNumber);
3407 request_val = (struct afs_request_val *)g_hash_table_lookup(
3408 afs_request_hash, &request_key);
3410 /* only allocate a new hash element when it's a request */
3412 if (!request_val && !reply)
3414 new_request_key = se_alloc(sizeof(struct afs_request_key));
3415 *new_request_key = request_key;
3417 request_val = se_alloc(sizeof(struct afs_request_val));
3418 request_val -> opcode = pntohl(&afsh->opcode);
3419 opcode = request_val->opcode;
3421 g_hash_table_insert(afs_request_hash, new_request_key,
3427 2.3 Dynamic conversation dissector registration.
3430 NOTE: This sections assumes that all information is available to
3431 create a complete conversation, source port/address and
3432 destination port/address. If either the destination port or
3433 address is know, see section 2.4 Dynamic server port dissector
3436 For protocols that negotiate a secondary port connection, for example
3437 packet-msproxy.c, a conversation can install a dissector to handle
3438 the secondary protocol dissection. After the conversation is created
3439 for the negotiated ports use the conversation_set_dissector to define
3440 the dissection routine.
3441 Before we create these conversations or assign a dissector to them we should
3442 first check that the conversation does not already exist and if it exists
3443 whether it is registered to our protocol or not.
3444 We should do this because it is uncommon but it does happen that multiple
3445 different protocols can use the same socketpair during different stages of
3446 an application cycle. By keeping track of the frame number a conversation
3447 was started in wireshark can still tell these different protocols apart.
3449 The second argument to conversation_set_dissector is a dissector handle,
3450 which is created with a call to create_dissector_handle or
3453 create_dissector_handle takes as arguments a pointer to the dissector
3454 function and a protocol ID as returned by proto_register_protocol;
3455 register_dissector takes as arguments a string giving a name for the
3456 dissector, a pointer to the dissector function, and a protocol ID.
3458 The protocol ID is the ID for the protocol dissected by the function.
3459 The function will not be called if the protocol has been disabled by the
3460 user; instead, the data for the protocol will be dissected as raw data.
3464 /* the handle for the dynamic dissector *
3465 static dissector_handle_t sub_dissector_handle;
3467 /* prototype for the dynamic dissector */
3468 static void sub_dissector(tvbuff_t *tvb, packet_info *pinfo,
3471 /* in the main protocol dissector, where the next dissector is setup */
3473 /* if conversation has a data field, create it and load structure */
3475 /* First check if a conversation already exists for this
3478 conversation = find_conversation(pinfo->fd->num,
3479 &pinfo->src, &pinfo->dst, protocol,
3480 src_port, dst_port, 0);
3482 /* If there is no such conversation, or if there is one but for
3483 someone else's protocol then we just create a new conversation
3484 and assign our protocol to it.
3486 if ( (conversation == NULL) ||
3487 (conversation->dissector_handle != sub_dissector_handle) ) {
3488 new_conv_info = se_alloc(sizeof(struct _new_conv_info));
3489 new_conv_info->data1 = value1;
3491 /* create the conversation for the dynamic port */
3492 conversation = conversation_new(pinfo->fd->num,
3493 &pinfo->src, &pinfo->dst, protocol,
3494 src_port, dst_port, new_conv_info, 0);
3496 /* set the dissector for the new conversation */
3497 conversation_set_dissector(conversation, sub_dissector_handle);
3502 proto_register_PROTOABBREV(void)
3506 sub_dissector_handle = create_dissector_handle(sub_dissector,
3512 2.4 Dynamic server port dissector registration.
3514 NOTE: While this example used both NO_ADDR2 and NO_PORT2 to create a
3515 conversation with only one port and address set, this isn't a
3516 requirement. Either the second port or the second address can be set
3517 when the conversation is created.
3519 For protocols that define a server address and port for a secondary
3520 protocol, a conversation can be used to link a protocol dissector to
3521 the server port and address. The key is to create the new
3522 conversation with the second address and port set to the "accept
3525 Some server applications can use the same port for different protocols during
3526 different stages of a transaction. For example it might initially use SNMP
3527 to perform some discovery and later switch to use TFTP using the same port.
3528 In order to handle this properly we must first check whether such a
3529 conversation already exists or not and if it exists we also check whether the
3530 registered dissector_handle for that conversation is "our" dissector or not.
3531 If not we create a new conversation on top of the previous one and set this new
3532 conversation to use our protocol.
3533 Since wireshark keeps track of the frame number where a conversation started
3534 wireshark will still be able to keep the packets apart even though they do use
3535 the same socketpair.
3536 (See packet-tftp.c and packet-snmp.c for examples of this)
3538 There are two support routines that will allow the second port and/or
3539 address to be set later.
3541 conversation_set_port2( conversation_t *conv, guint32 port);
3542 conversation_set_addr2( conversation_t *conv, address addr);
3544 These routines will change the second address or port for the
3545 conversation. So, the server port conversation will be converted into a
3546 more complete conversation definition. Don't use these routines if you
3547 want to create a conversation between the server and client and retain the
3548 server port definition, you must create a new conversation.
3553 /* the handle for the dynamic dissector *
3554 static dissector_handle_t sub_dissector_handle;
3558 /* in the main protocol dissector, where the next dissector is setup */
3560 /* if conversation has a data field, create it and load structure */
3562 new_conv_info = se_alloc(sizeof(struct _new_conv_info));
3563 new_conv_info->data1 = value1;
3565 /* create the conversation for the dynamic server address and port */
3566 /* NOTE: The second address and port values don't matter because the */
3567 /* NO_ADDR2 and NO_PORT2 options are set. */
3569 /* First check if a conversation already exists for this
3572 conversation = find_conversation(pinfo->fd->num,
3573 &server_src_addr, 0, protocol,
3574 server_src_port, 0, NO_ADDR2 | NO_PORT_B);
3575 /* If there is no such conversation, or if there is one but for
3576 someone else's protocol then we just create a new conversation
3577 and assign our protocol to it.
3579 if ( (conversation == NULL) ||
3580 (conversation->dissector_handle != sub_dissector_handle) ) {
3581 conversation = conversation_new(pinfo->fd->num,
3582 &server_src_addr, 0, protocol,
3583 server_src_port, 0, new_conv_info, NO_ADDR2 | NO_PORT2);
3585 /* set the dissector for the new conversation */
3586 conversation_set_dissector(conversation, sub_dissector_handle);
3589 2.5 Per-packet information.
3591 Information can be stored for each data packet that is processed by the
3592 dissector. The information is added with the p_add_proto_data function and
3593 retrieved with the p_get_proto_data function. The data pointers passed into
3594 the p_add_proto_data are not managed by the proto_data routines. If you use
3595 malloc or any other dynamic memory allocation scheme, you must release the
3596 data when it isn't required.
3599 p_add_proto_data(frame_data *fd, int proto, void *proto_data)
3601 p_get_proto_data(frame_data *fd, int proto)
3604 fd - The fd pointer in the pinfo structure, pinfo->fd
3605 proto - Protocol id returned by the proto_register_protocol call
3606 during initialization
3607 proto_data - pointer to the dissector data.
3610 2.6 User Preferences.
3612 If the dissector has user options, there is support for adding these preferences
3613 to a configuration dialog.
3615 You must register the module with the preferences routine with -
3617 module_t *prefs_register_protocol(proto_id, void (*apply_cb)(void))
3619 module_t *prefs_register_protocol_subtree(const char *subtree, int id,
3620 void (*apply_cb)(void));
3623 Where: proto_id - the value returned by "proto_register_protocol()" when
3624 the protocol was registered.
3625 apply_cb - Callback routine that is called when preferences are
3626 applied. It may be NULL, which inhibits the callback.
3627 subtree - grouping preferences tree node name (several protocols can
3628 be grouped under one preferences subtree)
3630 Then you can register the fields that can be configured by the user with these
3633 /* Register a preference with an unsigned integral value. */
3634 void prefs_register_uint_preference(module_t *module, const char *name,
3635 const char *title, const char *description, guint base, guint *var);
3637 /* Register a preference with an Boolean value. */
3638 void prefs_register_bool_preference(module_t *module, const char *name,
3639 const char *title, const char *description, gboolean *var);
3641 /* Register a preference with an enumerated value. */
3642 void prefs_register_enum_preference(module_t *module, const char *name,
3643 const char *title, const char *description, gint *var,
3644 const enum_val_t *enumvals, gboolean radio_buttons)
3646 /* Register a preference with a character-string value. */
3647 void prefs_register_string_preference(module_t *module, const char *name,
3648 const char *title, const char *description, char **var)
3650 /* Register a preference with a range of unsigned integers (e.g.,
3653 void prefs_register_range_preference(module_t *module, const char *name,
3654 const char *title, const char *description, range_t *var,
3657 Where: module - Returned by the prefs_register_protocol routine
3658 name - This is appended to the name of the protocol, with a
3659 "." between them, to construct a name that identifies
3660 the field in the preference file; the name itself
3661 should not include the protocol name, as the name in
3662 the preference file will already have it. Make sure that
3663 only lower-case ASCII letters, numbers, underscores and
3664 dots appear in the preference name.
3665 title - Field title in the preferences dialog
3666 description - Comments added to the preference file above the
3667 preference value and shown as tooltip in the GUI, or NULL
3668 var - pointer to the storage location that is updated when the
3669 field is changed in the preference dialog box. Note that
3670 with string preferences the given pointer is overwritten
3671 with a pointer to a new copy of the string during the
3672 preference registration. The passed-in string may be
3673 freed, but you must keep another pointer to the string
3674 in order to free it.
3675 base - Base that the unsigned integer is expected to be in,
3677 enumvals - an array of enum_val_t structures. This must be
3678 NULL-terminated; the members of that structure are:
3680 a short name, to be used with the "-o" flag - it
3681 should not contain spaces or upper-case letters,
3682 so that it's easier to put in a command line;
3684 a description, which is used in the GUI (and
3685 which, for compatibility reasons, is currently
3686 what's written to the preferences file) - it can
3687 contain spaces, capital letters, punctuation,
3690 the numerical value corresponding to that name
3692 radio_buttons - TRUE if the field is to be displayed in the
3693 preferences dialog as a set of radio buttons,
3694 FALSE if it is to be displayed as an option
3696 max_value - The maximum allowed value for a range (0 is the minimum).
3698 An example from packet-beep.c -
3700 proto_beep = proto_register_protocol("Blocks Extensible Exchange Protocol",
3705 /* Register our configuration options for BEEP, particularly our port */
3707 beep_module = prefs_register_protocol(proto_beep, proto_reg_handoff_beep);
3709 prefs_register_uint_preference(beep_module, "tcp.port", "BEEP TCP Port",
3710 "Set the port for BEEP messages (if other"
3711 " than the default of 10288)",
3712 10, &global_beep_tcp_port);
3714 prefs_register_bool_preference(beep_module, "strict_header_terminator",
3715 "BEEP Header Requires CRLF",
3716 "Specifies that BEEP requires CRLF as a "
3717 "terminator, and not just CR or LF",
3718 &global_beep_strict_term);
3720 This will create preferences "beep.tcp.port" and
3721 "beep.strict_header_terminator", the first of which is an unsigned
3722 integer and the second of which is a Boolean.
3724 Note that a warning will pop up if you've saved such preference to the
3725 preference file and you subsequently take the code out. The way to make
3726 a preference obsolete is to register it as such:
3728 /* Register a preference that used to be supported but no longer is. */
3729 void prefs_register_obsolete_preference(module_t *module,
3732 2.7 Reassembly/desegmentation for protocols running atop TCP.
3734 There are two main ways of reassembling a Protocol Data Unit (PDU) which
3735 spans across multiple TCP segments. The first approach is simpler, but
3736 assumes you are running atop of TCP when this occurs (but your dissector
3737 might run atop of UDP, too, for example), and that your PDUs consist of a
3738 fixed amount of data that includes enough information to determine the PDU
3739 length, possibly followed by additional data. The second method is more
3740 generic but requires more code and is less efficient.
3742 2.7.1 Using tcp_dissect_pdus().
3744 For the first method, you register two different dissection methods, one
3745 for the TCP case, and one for the other cases. It is a good idea to
3746 also have a dissect_PROTO_common function which will parse the generic
3747 content that you can find in all PDUs which is called from
3748 dissect_PROTO_tcp when the reassembly is complete and from
3749 dissect_PROTO_udp (or dissect_PROTO_other).
3751 To register the distinct dissector functions, consider the following
3752 example, stolen from packet-dns.c:
3754 dissector_handle_t dns_udp_handle;
3755 dissector_handle_t dns_tcp_handle;
3756 dissector_handle_t mdns_udp_handle;
3758 dns_udp_handle = create_dissector_handle(dissect_dns_udp,
3760 dns_tcp_handle = create_dissector_handle(dissect_dns_tcp,
3762 mdns_udp_handle = create_dissector_handle(dissect_mdns_udp,
3765 dissector_add_uint("udp.port", UDP_PORT_DNS, dns_udp_handle);
3766 dissector_add_uint("tcp.port", TCP_PORT_DNS, dns_tcp_handle);
3767 dissector_add_uint("udp.port", UDP_PORT_MDNS, mdns_udp_handle);
3768 dissector_add_uint("tcp.port", TCP_PORT_MDNS, dns_tcp_handle);
3770 The dissect_dns_udp function does very little work and calls
3771 dissect_dns_common, while dissect_dns_tcp calls tcp_dissect_pdus with a
3772 reference to a callback which will be called with reassembled data:
3775 dissect_dns_tcp(tvbuff_t *tvb, packet_info *pinfo, proto_tree *tree)
3777 tcp_dissect_pdus(tvb, pinfo, tree, dns_desegment, 2,
3778 get_dns_pdu_len, dissect_dns_tcp_pdu);
3781 (The dissect_dns_tcp_pdu function acts similarly to dissect_dns_udp.)
3782 The arguments to tcp_dissect_pdus are:
3784 the tvbuff pointer, packet_info pointer, and proto_tree pointer
3785 passed to the dissector;
3787 a gboolean flag indicating whether desegmentation is enabled for
3790 the number of bytes of PDU data required to determine the length
3793 a routine that takes as arguments a packet_info pointer, a tvbuff
3794 pointer and an offset value representing the offset into the tvbuff
3795 at which a PDU begins and should return - *without* throwing an
3796 exception (it is guaranteed that the number of bytes specified by the
3797 previous argument to tcp_dissect_pdus is available, but more data
3798 might not be available, so don't refer to any data past that) - the
3799 total length of the PDU, in bytes;
3801 a routine that's passed a tvbuff pointer, packet_info pointer,
3802 and proto_tree pointer, with the tvbuff containing a
3803 possibly-reassembled PDU, and that should dissect that PDU.
3805 2.7.2 Modifying the pinfo struct.
3807 The second reassembly mode is preferred when the dissector cannot determine
3808 how many bytes it will need to read in order to determine the size of a PDU.
3809 It may also be useful if your dissector needs to support reassembly from
3810 protocols other than TCP.
3812 Your dissect_PROTO will initially be passed a tvbuff containing the payload of
3813 the first packet. It should dissect as much data as it can, noting that it may
3814 contain more than one complete PDU. If the end of the provided tvbuff coincides
3815 with the end of a PDU then all is well and your dissector can just return as
3816 normal. (If it is a new-style dissector, it should return the number of bytes
3817 successfully processed.)
3819 If the dissector discovers that the end of the tvbuff does /not/ coincide with
3820 the end of a PDU, (ie, there is half of a PDU at the end of the tvbuff), it can
3821 indicate this to the parent dissector, by updating the pinfo struct. The
3822 desegment_offset field is the offset in the tvbuff at which the dissector will
3823 continue processing when next called. The desegment_len field should contain
3824 the estimated number of additional bytes required for completing the PDU. Next
3825 time your dissect_PROTO is called, it will be passed a tvbuff composed of the
3826 end of the data from the previous tvbuff together with desegment_len more bytes.
3828 If the dissector cannot tell how many more bytes it will need, it should set
3829 desegment_len=DESEGMENT_ONE_MORE_SEGMENT; it will then be called again as soon
3830 as any more data becomes available. Dissectors should set the desegment_len to a
3831 reasonable value when possible rather than always setting
3832 DESEGMENT_ONE_MORE_SEGMENT as it will generally be more efficient. Also, you
3833 *must not* set desegment_len=1 in this case, in the hope that you can change
3834 your mind later: once you return a positive value from desegment_len, your PDU
3835 boundary is set in stone.
3837 static hf_register_info hf[] = {
3839 {"C String", "c.string", FT_STRING, BASE_NONE, NULL, 0x0,
3845 * Dissect a buffer containing C strings.
3847 * @param tvb The buffer to dissect.
3848 * @param pinfo Packet Info.
3849 * @param tree The protocol tree.
3851 static void dissect_cstr(tvbuff_t * tvb, packet_info * pinfo, proto_tree * tree)
3854 while(offset < tvb_reported_length(tvb)) {
3855 gint available = tvb_reported_length_remaining(tvb, offset);
3856 gint len = tvb_strnlen(tvb, offset, available);
3859 /* we ran out of data: ask for more */
3860 pinfo->desegment_offset = offset;
3861 pinfo->desegment_len = DESEGMENT_ONE_MORE_SEGMENT;
3865 col_set_str(pinfo->cinfo, COL_INFO, "C String");
3867 len += 1; /* Add one for the '\0' */
3870 proto_tree_add_item(tree, hf_cstring, tvb, offset, len,
3873 offset += (guint)len;
3876 /* if we get here, then the end of the tvb coincided with the end of a
3877 string. Happy days. */
3880 This simple dissector will repeatedly return DESEGMENT_ONE_MORE_SEGMENT
3881 requesting more data until the tvbuff contains a complete C string. The C string
3882 will then be added to the protocol tree. Note that there may be more
3883 than one complete C string in the tvbuff, so the dissection is done in a
3888 The ptvcursor API allows a simpler approach to writing dissectors for
3889 simple protocols. The ptvcursor API works best for protocols whose fields
3890 are static and whose format does not depend on the value of other fields.
3891 However, even if only a portion of your protocol is statically defined,
3892 then that portion could make use of ptvcursors.
3894 The ptvcursor API lets you extract data from a tvbuff, and add it to a
3895 protocol tree in one step. It also keeps track of the position in the
3896 tvbuff so that you can extract data again without having to compute any
3897 offsets --- hence the "cursor" name of the API.
3899 The three steps for a simple protocol are:
3900 1. Create a new ptvcursor with ptvcursor_new()
3901 2. Add fields with multiple calls of ptvcursor_add()
3902 3. Delete the ptvcursor with ptvcursor_free()
3904 ptvcursor offers the possibility to add subtrees in the tree as well. It can be
3905 done in very simple steps :
3906 1. Create a new subtree with ptvcursor_push_subtree(). The old subtree is
3907 pushed in a stack and the new subtree will be used by ptvcursor.
3908 2. Add fields with multiple calls of ptvcursor_add(). The fields will be
3909 added in the new subtree created at the previous step.
3910 3. Pop the previous subtree with ptvcursor_pop_subtree(). The previous
3911 subtree is again used by ptvcursor.
3912 Note that at the end of the parsing of a packet you must have popped each
3913 subtree you pushed. If it's not the case, the dissector will generate an error.
3915 To use the ptvcursor API, include the "ptvcursor.h" file. The PGM dissector
3916 is an example of how to use it. You don't need to look at it as a guide;
3917 instead, the API description here should be good enough.
3919 2.8.1 ptvcursor API.
3922 ptvcursor_new(proto_tree* tree, tvbuff_t* tvb, gint offset)
3923 This creates a new ptvcursor_t object for iterating over a tvbuff.
3924 You must call this and use this ptvcursor_t object so you can use the
3928 ptvcursor_add(ptvcursor_t* ptvc, int hf, gint length, const guint encoding)
3929 This will extract 'length' bytes from the tvbuff and place it in
3930 the proto_tree as field 'hf', which is a registered header_field. The
3931 pointer to the proto_item that is created is passed back to you. Internally,
3932 the ptvcursor advances its cursor so the next call to ptvcursor_add
3933 starts where this call finished. The 'encoding' parameter is relevant for
3934 certain type of fields (See above under proto_tree_add_item()).
3937 ptvcursor_add_no_advance(ptvcursor_t* ptvc, int hf, gint length, const guint encoding)
3938 Like ptvcursor_add, but does not advance the internal cursor.
3941 ptvcursor_advance(ptvcursor_t* ptvc, gint length)
3942 Advances the internal cursor without adding anything to the proto_tree.
3945 ptvcursor_free(ptvcursor_t* ptvc)
3946 Frees the memory associated with the ptvcursor. You must call this
3947 after your dissection with the ptvcursor API is completed.
3951 ptvcursor_push_subtree(ptvcursor_t* ptvc, proto_item* it, gint ett_subtree)
3952 Pushes the current subtree in the tree stack of the cursor, creates a new
3953 one and sets this one as the working tree.
3956 ptvcursor_pop_subtree(ptvcursor_t* ptvc);
3957 Pops a subtree in the tree stack of the cursor
3960 ptvcursor_add_with_subtree(ptvcursor_t* ptvc, int hfindex, gint length,
3961 const guint encoding, gint ett_subtree);
3962 Adds an item to the tree and creates a subtree.
3963 If the length is unknown, length may be defined as SUBTREE_UNDEFINED_LENGTH.
3964 In this case, at the next pop, the item length will be equal to the advancement
3965 of the cursor since the creation of the subtree.
3968 ptvcursor_add_text_with_subtree(ptvcursor_t* ptvc, gint length,
3969 gint ett_subtree, const char* format, ...);
3970 Add a text node to the tree and create a subtree.
3971 If the length is unknown, length may be defined as SUBTREE_UNDEFINED_LENGTH.
3972 In this case, at the next pop, the item length will be equal to the advancement
3973 of the cursor since the creation of the subtree.
3975 2.8.2 Miscellaneous functions.
3978 ptvcursor_tvbuff(ptvcursor_t* ptvc)
3979 Returns the tvbuff associated with the ptvcursor.
3982 ptvcursor_current_offset(ptvcursor_t* ptvc)
3983 Returns the current offset.
3986 ptvcursor_tree(ptvcursor_t* ptvc)
3987 Returns the proto_tree associated with the ptvcursor.
3990 ptvcursor_set_tree(ptvcursor_t* ptvc, proto_tree *tree)
3991 Sets a new proto_tree for the ptvcursor.
3994 ptvcursor_set_subtree(ptvcursor_t* ptvc, proto_item* it, gint ett_subtree);
3995 Creates a subtree and adds it to the cursor as the working tree but does
3996 not save the old working tree.